commit from root

Baichuan-13B-Chat-full/all_results.json

@@ -1,7 +0,0 @@
-{
-    "epoch": 2.0,
-    "train_loss": 0.5708327819994277,
-    "train_runtime": 105327.89,
-    "train_samples_per_second": 4.797,
-    "train_steps_per_second": 0.019
-}

Baichuan-13B-Chat-full/config.json

@@ -1,29 +0,0 @@
-{
-  "_from_model_config": true,
-  "_name_or_path": "baichuan-inc/Baichuan-13B-Chat",
-  "architectures": [
-    "BaichuanForCausalLM"
-  ],
-  "auto_map": {
-    "AutoConfig": "configuration_baichuan.BaichuanConfig",
-    "AutoModel": "modeling_baichuan.BaichuanForCausalLM",
-    "AutoModelForCausalLM": "baichuan-inc/Baichuan-13B-Chat--modeling_baichuan.BaichuanForCausalLM"
-  },
-  "bos_token_id": 1,
-  "eos_token_id": 2,
-  "hidden_act": "silu",
-  "hidden_size": 5120,
-  "initializer_range": 0.02,
-  "intermediate_size": 13696,
-  "model_max_length": 4096,
-  "model_type": "baichuan",
-  "num_attention_heads": 40,
-  "num_hidden_layers": 40,
-  "pad_token_id": 0,
-  "rms_norm_eps": 1e-06,
-  "tie_word_embeddings": false,
-  "torch_dtype": "float16",
-  "transformers_version": "4.31.0",
-  "use_cache": false,
-  "vocab_size": 64000
-}

Baichuan-13B-Chat-full/configuration_baichuan.py

@@ -1,46 +0,0 @@
-# Copyright (c) 2023, Baichuan Intelligent Technology. All rights reserved.
-
-from transformers.configuration_utils import PretrainedConfig
-
-class BaichuanConfig(PretrainedConfig):
-    model_type = "baichuan"
-    keys_to_ignore_at_inference = ["past_key_values"]
-
-    def __init__(
-        self,
-        vocab_size=64000,
-        hidden_size=5120,
-        intermediate_size=13696,
-        num_hidden_layers=40,
-        num_attention_heads=40,
-        hidden_act="silu",
-        model_max_length=4096,
-        initializer_range=0.02,
-        rms_norm_eps=1e-6,
-        use_cache=True,
-        pad_token_id=0,
-        bos_token_id=1,
-        eos_token_id=2,
-        tie_word_embeddings=False,
-        gradient_checkpointing=False,
-        **kwargs,
-    ):
-        self.vocab_size = vocab_size
-        self.model_max_length = model_max_length
-        self.hidden_size = hidden_size
-        self.intermediate_size = intermediate_size
-        self.num_hidden_layers = num_hidden_layers
-        self.num_attention_heads = num_attention_heads
-        self.hidden_act = hidden_act
-        self.initializer_range = initializer_range
-        self.rms_norm_eps = rms_norm_eps
-        self.use_cache = use_cache
-        self.gradient_checkpointing = gradient_checkpointing,
-        super().__init__(
-            pad_token_id=pad_token_id,
-            bos_token_id=bos_token_id,
-            eos_token_id=eos_token_id,
-            tie_word_embeddings=tie_word_embeddings,
-            **kwargs,
-        )
-

Baichuan-13B-Chat-full/generation_config.json

@@ -1,14 +0,0 @@
-{
-  "assistant_token_id": 196,
-  "bos_token_id": 1,
-  "do_sample": true,
-  "eos_token_id": 2,
-  "max_new_tokens": 2048,
-  "pad_token_id": 0,
-  "repetition_penalty": 1.1,
-  "temperature": 0.3,
-  "top_k": 5,
-  "top_p": 0.85,
-  "transformers_version": "4.31.0",
-  "user_token_id": 195
-}

Baichuan-13B-Chat-full/generation_utils.py

@@ -1,82 +0,0 @@
-from typing import List
-from queue import Queue
-
-import torch
-
-
-def build_chat_input(model, tokenizer, messages: List[dict], max_new_tokens: int=0):
-    def _parse_messages(messages, split_role="user"):
-        system, rounds = "", []
-        round = []
-        for i, message in enumerate(messages):
-            if message["role"] == "system":
-                assert i == 0
-                system = message["content"]
-                continue
-            if message["role"] == split_role and round:
-                rounds.append(round)
-                round = []
-            round.append(message)
-        if round:
-            rounds.append(round)
-        return system, rounds
-
-    max_new_tokens = max_new_tokens or model.generation_config.max_new_tokens
-    max_input_tokens = model.config.model_max_length - max_new_tokens
-    system, rounds = _parse_messages(messages, split_role="user")
-    system_tokens = tokenizer.encode(system)
-    max_history_tokens = max_input_tokens - len(system_tokens)
-
-    history_tokens = []
-    for round in rounds[::-1]:
-        round_tokens = []
-        for message in round:
-            if message["role"] == "user":
-                round_tokens.append(model.generation_config.user_token_id)
-            else:
-                round_tokens.append(model.generation_config.assistant_token_id)
-            round_tokens.extend(tokenizer.encode(message["content"]))
-        if len(history_tokens) == 0 or len(history_tokens) + len(round_tokens) <= max_history_tokens:
-            history_tokens = round_tokens + history_tokens  # concat left
-            if len(history_tokens) < max_history_tokens:
-                continue
-        break
-
-    input_tokens = system_tokens + history_tokens
-    if messages[-1]["role"] != "assistant":
-        input_tokens.append(model.generation_config.assistant_token_id)
-    input_tokens = input_tokens[-max_input_tokens:]  # truncate left
-    return torch.LongTensor([input_tokens]).to(model.device)
-
-
-class TextIterStreamer:
-    def __init__(self, tokenizer, skip_prompt=False, skip_special_tokens=False):
-        self.tokenizer = tokenizer
-        self.skip_prompt = skip_prompt
-        self.skip_special_tokens = skip_special_tokens
-        self.tokens = []
-        self.text_queue = Queue()
-        self.next_tokens_are_prompt = True
-
-    def put(self, value):
-        if self.skip_prompt and self.next_tokens_are_prompt:
-            self.next_tokens_are_prompt = False
-        else:
-            if len(value.shape) > 1:
-                value = value[0]
-            self.tokens.extend(value.tolist())
-            self.text_queue.put(
-                self.tokenizer.decode(self.tokens, skip_special_tokens=self.skip_special_tokens))
-
-    def end(self):
-        self.text_queue.put(None)
-
-    def __iter__(self):
-        return self
-
-    def __next__(self):
-        value = self.text_queue.get()
-        if value is None:
-            raise StopIteration()
-        else:
-            return value

Baichuan-13B-Chat-full/modeling_baichuan.py

@@ -1,572 +0,0 @@
-# Copyright (c) 2023, Baichuan Intelligent Technology. All rights reserved.
-
-import math
-from threading import Thread
-from typing import List, Optional, Tuple, Union
-
-import torch
-import torch.utils.checkpoint
-from torch.nn import CrossEntropyLoss
-from transformers import PreTrainedModel
-from transformers.activations import ACT2FN
-from transformers.modeling_outputs import BaseModelOutputWithPast, CausalLMOutputWithPast
-from transformers.utils import logging
-from transformers.generation.utils import GenerationConfig
-
-from .configuration_baichuan import BaichuanConfig
-from .generation_utils import build_chat_input, TextIterStreamer
-
-logger = logging.get_logger(__name__)
-
-
-def _get_interleave(n):
-    def _get_interleave_power_of_2(n):
-        start = (2 ** (-2 ** -(math.log2(n) - 3)))
-        ratio = start
-        return [start * ratio ** i for i in range(n)]
-
-    if math.log2(n).is_integer():
-        return _get_interleave_power_of_2(n)
-    else:
-        closest_power_of_2 = 2 ** math.floor(math.log2(n))
-        return _get_interleave_power_of_2(closest_power_of_2) + \
-               _get_interleave(2 * closest_power_of_2)[0::2][:n - closest_power_of_2]
-
-def _fill_with_neg_inf(t):
-    """FP16-compatible function that fills a tensor with -inf."""
-    return t.float().fill_(float("-inf")).type_as(t)
-
-def _gen_alibi_mask(n_head, max_pos):
-    """used in inference only"""
-    slopes = torch.Tensor(_get_interleave(n_head))
-    alibi = slopes.unsqueeze(1).unsqueeze(1) * torch.arange(max_pos).unsqueeze(0).unsqueeze(0).expand(
-        n_head, -1, -1)
-    alibi = alibi.view(n_head, 1, max_pos)
-    alibi_mask = torch.triu(
-        _fill_with_neg_inf(torch.zeros([max_pos, max_pos])), 1
-    )
-    alibi_mask = alibi_mask.unsqueeze(0) + alibi
-    return alibi_mask
-
-def _buffered_future_mask(tensor, maxpos, alibi, attn_heads):
-    """used in training only"""
-    dim = tensor.size(1)
-    _future_mask = torch.triu(
-        _fill_with_neg_inf(torch.zeros([maxpos, maxpos])), 1
-    )   
-    _future_mask = _future_mask.unsqueeze(0) + alibi
-    _future_mask = _future_mask.to(tensor)
-    return _future_mask[:tensor.shape[0] * attn_heads, :maxpos, :maxpos]
-
-
-class RMSNorm(torch.nn.Module):
-    def __init__(self, hidden_size, epsilon=1e-6):
-        super().__init__()
-        self.weight = torch.nn.Parameter(torch.empty(hidden_size))
-        self.epsilon = epsilon
-
-    def forward(self, hidden_states):
-        variance = hidden_states.to(torch.float32).pow(2).mean(-1, keepdim=True)
-        hidden_states = hidden_states * torch.rsqrt(variance + self.epsilon)
-
-        # convert into half-precision
-        if self.weight.dtype in [torch.float16, torch.bfloat16]:
-            hidden_states = hidden_states.to(self.weight.dtype)
-
-        return self.weight * hidden_states
-
-
-class MLP(torch.nn.Module):
-    def __init__(
-            self,
-            hidden_size: int,
-            intermediate_size: int,
-            hidden_act: str,
-    ):
-        super().__init__()
-        self.gate_proj = torch.nn.Linear(hidden_size, intermediate_size, bias=False)
-        self.down_proj = torch.nn.Linear(intermediate_size, hidden_size, bias=False)
-        self.up_proj = torch.nn.Linear(hidden_size, intermediate_size, bias=False)
-        self.act_fn = ACT2FN[hidden_act]
-
-    def forward(self, x):
-        return self.down_proj(self.act_fn(self.gate_proj(x)) * self.up_proj(x))
-
-
-class BaichuanAttention(torch.nn.Module):
-    def __init__(self, config: BaichuanConfig):
-        super().__init__()
-        self.config = config
-        self.hidden_size = config.hidden_size
-        self.num_heads = config.num_attention_heads
-        self.head_dim = self.hidden_size // self.num_heads
-        self.max_position_embeddings = config.model_max_length
-
-        if (self.head_dim * self.num_heads) != self.hidden_size:
-            raise ValueError(
-                f"hidden_size {self.hidden_size} is not divisible by num_heads {self.num_heads}"
-            )
-        self.W_pack = torch.nn.Linear(self.hidden_size, 3 * self.hidden_size, bias=False)
-        self.o_proj = torch.nn.Linear(self.num_heads * self.head_dim, self.hidden_size, bias=False)
-
-    def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
-        return tensor.view(bsz, seq_len, self.num_heads, self.head_dim).transpose(1, 2).contiguous()
-
-    def forward(
-            self,
-            hidden_states: torch.Tensor,
-            attention_mask: Optional[torch.Tensor] = None,
-            past_key_value: Optional[Tuple[torch.Tensor]] = None,
-            output_attentions: bool = False,
-            use_cache: bool = False,
-    ) -> Tuple[torch.Tensor, Optional[torch.Tensor], Optional[Tuple[torch.Tensor]]]:
-
-        bsz, q_len, _ = hidden_states.size()
-
-        proj = self.W_pack(hidden_states)
-        proj = proj.unflatten(-1, (3, self.hidden_size)).unsqueeze(0).transpose(0, -2).squeeze(-2)
-        query_states = proj[0].view(bsz, q_len, self.num_heads, self.head_dim).transpose(1, 2)
-        key_states = proj[1].view(bsz, q_len, self.num_heads, self.head_dim).transpose(1, 2)
-        value_states = proj[2].view(bsz, q_len, self.num_heads, self.head_dim).transpose(1, 2)
-
-        kv_seq_len = key_states.shape[-2]
-        if past_key_value is not None:
-            kv_seq_len += past_key_value[0].shape[-2]
-
-        if past_key_value is not None:
-            # reuse k, v, self_attention
-            key_states = torch.cat([past_key_value[0], key_states], dim=2)
-            value_states = torch.cat([past_key_value[1], value_states], dim=2)
-
-        past_key_value = (key_states, value_states) if use_cache else None
-
-        attn_weights = torch.matmul(query_states, key_states.transpose(2, 3)) / math.sqrt(self.head_dim)
-
-        if attention_mask is not None:
-            if q_len == 1: # inference with cache
-                if len(attention_mask.size()) == 4:
-                    attention_mask = attention_mask[:, :, -1:, :]   
-                else:
-                    attention_mask = attention_mask[:, -1:, :]    
-            attn_weights = attn_weights + attention_mask
-            attn_weights = torch.max(attn_weights, torch.tensor(torch.finfo(attn_weights.dtype).min))
-
-        attn_weights = torch.nn.functional.softmax(attn_weights, dim=-1)
-
-        attn_output = torch.matmul(attn_weights, value_states)
-
-        attn_output = attn_output.transpose(1, 2)
-        attn_output = attn_output.reshape(bsz, q_len, self.hidden_size)
-        attn_output = self.o_proj(attn_output)
-
-        if not output_attentions:
-            attn_weights = None
-
-        return attn_output, attn_weights, past_key_value
-
-
-class BaichuanLayer(torch.nn.Module):
-    def __init__(self, config: BaichuanConfig):
-        super().__init__()
-        self.hidden_size = config.hidden_size
-        self.self_attn = BaichuanAttention(config=config)
-        self.mlp = MLP(
-            hidden_size=self.hidden_size,
-            intermediate_size=config.intermediate_size,
-            hidden_act=config.hidden_act,
-        )
-        self.input_layernorm = RMSNorm(config.hidden_size, epsilon=config.rms_norm_eps)
-        self.post_attention_layernorm = RMSNorm(config.hidden_size, epsilon=config.rms_norm_eps)
-
-    def forward(
-            self,
-            hidden_states: torch.Tensor,
-            attention_mask: Optional[torch.Tensor] = None,
-            past_key_value: Optional[Tuple[torch.Tensor]] = None,
-            output_attentions: Optional[bool] = False,
-            use_cache: Optional[bool] = False,
-    ) -> Tuple[torch.FloatTensor, Optional[Tuple[torch.FloatTensor, torch.FloatTensor]]]:
-
-        residual = hidden_states
-
-        hidden_states = self.input_layernorm(hidden_states)
-
-        # Self Attention
-        hidden_states, self_attn_weights, present_key_value = self.self_attn(
-            hidden_states=hidden_states,
-            attention_mask=attention_mask,
-            past_key_value=past_key_value,
-            output_attentions=output_attentions,
-            use_cache=use_cache,
-        )
-        hidden_states = residual + hidden_states
-
-        # Fully Connected
-        residual = hidden_states
-        hidden_states = self.post_attention_layernorm(hidden_states)
-        hidden_states = self.mlp(hidden_states)
-        hidden_states = residual + hidden_states
-
-        outputs = (hidden_states,)
-
-        if use_cache:
-            outputs += (present_key_value,)
-
-        return outputs
-
-
-class BaichuanPreTrainedModel(PreTrainedModel):
-    config_class = BaichuanConfig
-    base_model_prefix = "model"
-    supports_gradient_checkpointing = True
-    _no_split_modules = ["BaichuanLayer"]
-    _keys_to_ignore_on_load_unexpected = [r"decoder\.version"]
-
-    def _init_weights(self, module):
-        std = self.config.initializer_range
-        if isinstance(module, torch.nn.Linear):
-            module.weight.data.normal_(mean=0.0, std=std)
-            if module.bias is not None:
-                module.bias.data.zero_()
-        elif isinstance(module, torch.nn.Embedding):
-            module.weight.data.normal_(mean=0.0, std=std)
-            if module.padding_idx is not None:
-                module.weight.data[module.padding_idx].zero_()
-
-    def _set_gradient_checkpointing(self, module, value=False):
-        if isinstance(module, BaichuanModel):
-            module.gradient_checkpointing = value
-
-
-class BaichuanModel(BaichuanPreTrainedModel):
-    def __init__(self, config: BaichuanConfig):
-        super().__init__(config)
-        self.padding_idx = config.pad_token_id
-        self.vocab_size = config.vocab_size
-        self.n_head = config.num_attention_heads
-        self.embed_tokens = torch.nn.Embedding(config.vocab_size, config.hidden_size, self.padding_idx)
-        self.layers = torch.nn.ModuleList([BaichuanLayer(config) for _ in range(config.num_hidden_layers)])
-        self.norm = RMSNorm(config.hidden_size, epsilon=config.rms_norm_eps)
-
-        self.gradient_checkpointing = config.gradient_checkpointing
-        self.post_init()
-        self.max_cache_pos = config.model_max_length
-        self.first_run = True
-        self.alibi_mask = None
-
-    def get_input_embeddings(self):
-        return self.embed_tokens
-
-    def set_input_embeddings(self, value):
-        self.embed_tokens = value
-
-    def get_alibi_mask(self, tensor, seq_length_with_past):
-        if self.training:
-            slopes = torch.Tensor(_get_interleave(self.n_head))
-            alibi = slopes.unsqueeze(1).unsqueeze(1) * torch.arange(seq_length_with_past).unsqueeze(0).unsqueeze(0).expand(
-                self.n_head,
-                -1, -1) 
-            alibi = alibi.view(self.n_head, 1, seq_length_with_past)
-            mask = _buffered_future_mask(tensor, seq_length_with_past, alibi, self.n_head)
-        else:
-            if self.first_run:
-                self.first_run = False
-                self.register_buffer("future_mask", _gen_alibi_mask(self.n_head, self.max_cache_pos).to(tensor), persistent=False)
-            if seq_length_with_past > self.max_cache_pos:
-                self.max_cache_pos = seq_length_with_past
-                self.register_buffer("future_mask", _gen_alibi_mask(self.n_head, self.max_cache_pos).to(tensor), persistent=False)
-            mask = self.future_mask[:self.n_head, :seq_length_with_past, :seq_length_with_past] 
-        return mask
-
-    def forward(
-            self,
-            input_ids: torch.LongTensor = None,
-            attention_mask: Optional[torch.Tensor] = None,
-            past_key_values: Optional[List[torch.FloatTensor]] = None,
-            inputs_embeds: Optional[torch.FloatTensor] = None,
-            use_cache: Optional[bool] = False,
-            output_attentions: Optional[bool] = False,
-            output_hidden_states: Optional[bool] = False,
-            return_dict: Optional[bool] = True,
-    ) -> Union[Tuple, BaseModelOutputWithPast]:
-
-        if input_ids is not None and inputs_embeds is not None:
-            raise ValueError("You cannot provide both input_ids and inputs_embeds simultaneously")
-        elif input_ids is not None:
-            batch_size, seq_length = input_ids.shape
-        elif inputs_embeds is not None:
-            batch_size, seq_length, _ = inputs_embeds.shape
-        else:
-            raise ValueError("You need to provide input_ids or inputs_embeds")
-
-        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
-
-        seq_length_with_past = seq_length
-
-        if past_key_values is not None:
-            past_key_values_length = past_key_values[0][0].shape[2]
-            seq_length_with_past = seq_length_with_past + past_key_values_length
-
-        if inputs_embeds is None:
-            inputs_embeds = self.embed_tokens(input_ids)
-
-        if self.training:
-            if self.alibi_mask is None or self.alibi_mask.shape[-1] != seq_length_with_past:
-                self.alibi_mask = self.get_alibi_mask(inputs_embeds, seq_length_with_past)
-            alibi_mask = self.alibi_mask
-        else:
-            alibi_mask = self.get_alibi_mask(inputs_embeds, seq_length_with_past)
-
-        if attention_mask is not None:
-            if len(attention_mask.shape) == 2:
-                expanded_mask = attention_mask.to(alibi_mask.dtype)
-                expanded_mask = torch.tril(torch.gt(expanded_mask[:, :, None] * expanded_mask[:, None, :], 0)
-                                ) * torch.eq(expanded_mask[:, :, None] - expanded_mask[:, None, :], 0)
-            else:
-                expanded_mask = attention_mask 
-            bsz = inputs_embeds.size(0)
-            src_len, tgt_len = alibi_mask.size()[-2:]
-            expanded_mask = expanded_mask.unsqueeze(1).expand(bsz, 1, src_len, tgt_len).to(alibi_mask.dtype)
-            inverted_mask = 1.0 - expanded_mask
-            inverted_mask = inverted_mask.masked_fill(inverted_mask.to(torch.bool), torch.finfo(alibi_mask.dtype).min)
-            attention_mask = inverted_mask + alibi_mask.unsqueeze(0)
-        else:
-            attention_mask = alibi_mask
-
-        hidden_states = inputs_embeds
-
-        if self.gradient_checkpointing and self.training:
-            if use_cache:
-                logger.warning_once(
-                    "`use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`..."
-                )
-                use_cache = False
-
-        # decoder layers
-        all_hidden_states = () if output_hidden_states else None
-        all_self_attns = () if output_attentions else None
-        next_decoder_cache = () if use_cache else None
-
-        for idx, decoder_layer in enumerate(self.layers):
-            if output_hidden_states:
-                all_hidden_states += (hidden_states,)
-
-            past_key_value = past_key_values[idx] if past_key_values is not None else None
-
-            if self.gradient_checkpointing and self.training:
-
-                def create_custom_forward(module):
-                    def custom_forward(*inputs):
-                        # None for past_key_value
-                        return module(*inputs, output_attentions, None)
-
-                    return custom_forward
-
-                layer_outputs = torch.utils.checkpoint.checkpoint(
-                    create_custom_forward(decoder_layer),
-                    hidden_states,
-                    attention_mask,
-                    None,
-                )
-            else:
-                layer_outputs = decoder_layer(
-                    hidden_states,
-                    attention_mask=attention_mask,
-                    past_key_value=past_key_value,
-                    output_attentions=output_attentions,
-                    use_cache=use_cache,
-                )
-
-            hidden_states = layer_outputs[0]
-
-            if use_cache:
-                next_decoder_cache += (layer_outputs[2 if output_attentions else 1],)
-
-            if output_attentions:
-                all_self_attns += (layer_outputs[1],)
-
-        hidden_states = self.norm(hidden_states)
-
-        # add hidden states from the last decoder layer
-        if output_hidden_states:
-            all_hidden_states += (hidden_states,)
-
-        next_cache = next_decoder_cache if use_cache else None
-        if not return_dict:
-            return tuple(v for v in [hidden_states, next_cache, all_hidden_states, all_self_attns] if v is not None)
-        return BaseModelOutputWithPast(
-            last_hidden_state=hidden_states,
-            past_key_values=next_cache,
-            hidden_states=all_hidden_states,
-            attentions=all_self_attns,
-        )
-
-
-class BaichuanForCausalLM(BaichuanPreTrainedModel):
-    def __init__(self, config):
-        super().__init__(config)
-        self.model = BaichuanModel(config)
-        self.lm_head = torch.nn.Linear(config.hidden_size, config.vocab_size, bias=False)
-
-        # Initialize weights and apply final processing
-        self.post_init()
-
-    def get_input_embeddings(self):
-        return self.model.embed_tokens
-
-    def set_input_embeddings(self, value):
-        self.model.embed_tokens = value
-
-    def get_output_embeddings(self):
-        return self.lm_head
-
-    def set_output_embeddings(self, new_embeddings):
-        self.lm_head = new_embeddings
-
-    def set_decoder(self, decoder):
-        self.model = decoder
-
-    def get_decoder(self):
-        return self.model
-
-    def forward(
-            self,
-            input_ids: torch.LongTensor = None,
-            attention_mask: Optional[torch.Tensor] = None,
-            past_key_values: Optional[List[torch.FloatTensor]] = None,
-            inputs_embeds: Optional[torch.FloatTensor] = None,
-            labels: Optional[torch.LongTensor] = None,
-            use_cache: Optional[bool] = None,
-            output_attentions: Optional[bool] = False,
-            output_hidden_states: Optional[bool] = False,
-            return_dict: Optional[bool] = True,
-            **kwargs
-    ) -> Union[Tuple, CausalLMOutputWithPast]:
-
-        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
-        
-        # decoder outputs consists of (dec_features, layer_state, dec_hidden, dec_attn)
-        outputs = self.model(
-            input_ids=input_ids,
-            attention_mask=attention_mask,
-            past_key_values=past_key_values,
-            inputs_embeds=inputs_embeds,
-            use_cache=use_cache,
-            output_attentions=output_attentions,
-            output_hidden_states=output_hidden_states,
-            return_dict=return_dict,
-        )
-
-        hidden_states = outputs[0]
-        logits = self.lm_head(hidden_states)
-
-        loss = None
-        if labels is not None:
-            # Shift so that tokens < n predict n
-            shift_logits = logits[..., :-1, :].contiguous()
-            shift_labels = labels[..., 1:].contiguous()
-            # Flatten the tokens
-            loss_fct = CrossEntropyLoss()
-            shift_logits = shift_logits.view(-1, self.config.vocab_size)
-            shift_labels = shift_labels.view(-1)
-            # Enable model parallelism
-            shift_labels = shift_labels.to(shift_logits.device)
-            loss = loss_fct(shift_logits, shift_labels)
-
-        if not return_dict:
-            output = (logits,) + outputs[1:]
-            return (loss,) + output if loss is not None else output
-
-        return CausalLMOutputWithPast(
-            loss=loss,
-            logits=logits,
-            past_key_values=outputs.past_key_values,
-            hidden_states=outputs.hidden_states,
-            attentions=outputs.attentions,
-        )
-
-    def prepare_inputs_for_generation(
-            self,
-            input_ids: torch.LongTensor,
-            past_key_values: Optional[torch.Tensor] = None,
-            attention_mask: Optional[torch.Tensor] = None,
-            inputs_embeds: Optional[torch.Tensor] = None,
-            **kwargs
-    ):
-        if past_key_values:
-            input_ids = input_ids[:, -1:]
-
-        # if `inputs_embeds` are passed, we only want to use them in the 1st generation step
-        if inputs_embeds is not None and past_key_values is None:
-            model_inputs = {"inputs_embeds": inputs_embeds}
-        else:
-            model_inputs = {"input_ids": input_ids}
-
-        model_inputs.update(
-            {
-                "past_key_values": past_key_values,
-                "use_cache": kwargs.get("use_cache"),
-                "attention_mask": attention_mask
-            }
-        )
-        return model_inputs
-
-    @staticmethod
-    def _reorder_cache(past_key_values, beam_idx):
-        return tuple(
-            tuple(past_state.index_select(0, beam_idx) for past_state in layer_past)
-            for layer_past in past_key_values
-        )
-
-    def quantize(self, bits: int):
-        try:
-            from .quantizer import QLinear
-        except ImportError:
-            raise ImportError(
-                f"Needs QLinear to run quantize."
-            )
-
-        for layer in self.model.layers:
-            layer.self_attn.W_pack = QLinear(
-                bits=bits,
-                weight=layer.self_attn.W_pack.weight,
-                bias = None,
-            )
-            layer.self_attn.o_proj = QLinear(
-                bits=bits,
-                weight=layer.self_attn.o_proj.weight,
-                bias = None,
-            )
-            layer.mlp.gate_proj = QLinear(
-                bits=bits,
-                weight=layer.mlp.gate_proj.weight,
-                bias = None,
-            )
-            layer.mlp.down_proj = QLinear(
-                bits=bits,
-                weight=layer.mlp.down_proj.weight,
-                bias = None,
-            )
-            layer.mlp.up_proj = QLinear(
-                bits=bits,
-                weight=layer.mlp.up_proj.weight,
-                bias = None,
-            )
-        return self
-
-    @torch.no_grad()
-    def chat(self, tokenizer, messages: List[dict], stream=False,
-             generation_config: Optional[GenerationConfig]=None):
-        generation_config = generation_config or self.generation_config
-        input_ids = build_chat_input(self, tokenizer, messages, generation_config.max_new_tokens)
-        if stream:
-            streamer = TextIterStreamer(tokenizer, skip_prompt=True, skip_special_tokens=True)
-            Thread(target=self.generate, kwargs=dict(
-                inputs=input_ids, streamer=streamer,
-                generation_config=generation_config,
-            )).start()
-            return streamer
-        else:
-            outputs = self.generate(input_ids, generation_config=generation_config)
-            response = tokenizer.decode(outputs[0][len(input_ids[0]):], skip_special_tokens=True)
-            return response

Baichuan-13B-Chat-full/pytorch_model-00002-of-00003.bin

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:f5e1e6afc5fd592dda61582a07f748ef813b21ec6885eba8fe2da9cf973931c4
-size 9947420311

Baichuan-13B-Chat-full/pytorch_model-00003-of-00003.bin

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:ee03255509602bdb0fc2cd144bd3df07fcd7ad2696778e11dfdf380751698ba1
-size 6610199816

Baichuan-13B-Chat-full/pytorch_model.bin.index.json

@@ -1,290 +0,0 @@
-{
-  "metadata": {
-    "total_size": 26529802240
-  },
-  "weight_map": {
-    "lm_head.weight": "pytorch_model-00003-of-00003.bin",
-    "model.embed_tokens.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.0.input_layernorm.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.0.mlp.down_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.0.mlp.gate_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.0.mlp.up_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.0.post_attention_layernorm.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.0.self_attn.W_pack.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.0.self_attn.o_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.1.input_layernorm.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.1.mlp.down_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.1.mlp.gate_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.1.mlp.up_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.1.post_attention_layernorm.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.1.self_attn.W_pack.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.1.self_attn.o_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.10.input_layernorm.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.10.mlp.down_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.10.mlp.gate_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.10.mlp.up_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.10.post_attention_layernorm.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.10.self_attn.W_pack.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.10.self_attn.o_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.11.input_layernorm.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.11.mlp.down_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.11.mlp.gate_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.11.mlp.up_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.11.post_attention_layernorm.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.11.self_attn.W_pack.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.11.self_attn.o_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.12.input_layernorm.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.12.mlp.down_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.12.mlp.gate_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.12.mlp.up_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.12.post_attention_layernorm.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.12.self_attn.W_pack.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.12.self_attn.o_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.13.input_layernorm.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.13.mlp.down_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.13.mlp.gate_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.13.mlp.up_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.13.post_attention_layernorm.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.13.self_attn.W_pack.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.13.self_attn.o_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.14.input_layernorm.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.14.mlp.down_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.14.mlp.gate_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.14.mlp.up_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.14.post_attention_layernorm.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.14.self_attn.W_pack.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.14.self_attn.o_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.15.input_layernorm.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.15.mlp.down_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.15.mlp.gate_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.15.mlp.up_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.15.post_attention_layernorm.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.15.self_attn.W_pack.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.15.self_attn.o_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.16.input_layernorm.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.16.mlp.down_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.16.mlp.gate_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.16.mlp.up_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.16.post_attention_layernorm.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.16.self_attn.W_pack.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.16.self_attn.o_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.17.input_layernorm.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.17.mlp.down_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.17.mlp.gate_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.17.mlp.up_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.17.post_attention_layernorm.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.17.self_attn.W_pack.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.17.self_attn.o_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.18.input_layernorm.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.18.mlp.down_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.18.mlp.gate_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.18.mlp.up_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.18.post_attention_layernorm.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.18.self_attn.W_pack.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.18.self_attn.o_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.19.input_layernorm.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.19.mlp.down_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.19.mlp.gate_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.19.mlp.up_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.19.post_attention_layernorm.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.19.self_attn.W_pack.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.19.self_attn.o_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.2.input_layernorm.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.2.mlp.down_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.2.mlp.gate_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.2.mlp.up_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.2.post_attention_layernorm.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.2.self_attn.W_pack.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.2.self_attn.o_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.20.input_layernorm.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.20.mlp.down_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.20.mlp.gate_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.20.mlp.up_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.20.post_attention_layernorm.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.20.self_attn.W_pack.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.20.self_attn.o_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.21.input_layernorm.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.21.mlp.down_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.21.mlp.gate_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.21.mlp.up_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.21.post_attention_layernorm.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.21.self_attn.W_pack.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.21.self_attn.o_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.22.input_layernorm.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.22.mlp.down_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.22.mlp.gate_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.22.mlp.up_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.22.post_attention_layernorm.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.22.self_attn.W_pack.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.22.self_attn.o_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.23.input_layernorm.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.23.mlp.down_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.23.mlp.gate_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.23.mlp.up_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.23.post_attention_layernorm.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.23.self_attn.W_pack.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.23.self_attn.o_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.24.input_layernorm.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.24.mlp.down_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.24.mlp.gate_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.24.mlp.up_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.24.post_attention_layernorm.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.24.self_attn.W_pack.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.24.self_attn.o_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.25.input_layernorm.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.25.mlp.down_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.25.mlp.gate_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.25.mlp.up_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.25.post_attention_layernorm.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.25.self_attn.W_pack.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.25.self_attn.o_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.26.input_layernorm.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.26.mlp.down_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.26.mlp.gate_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.26.mlp.up_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.26.post_attention_layernorm.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.26.self_attn.W_pack.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.26.self_attn.o_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.27.input_layernorm.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.27.mlp.down_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.27.mlp.gate_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.27.mlp.up_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.27.post_attention_layernorm.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.27.self_attn.W_pack.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.27.self_attn.o_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.28.input_layernorm.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.28.mlp.down_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.28.mlp.gate_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.28.mlp.up_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.28.post_attention_layernorm.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.28.self_attn.W_pack.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.28.self_attn.o_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.29.input_layernorm.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.29.mlp.down_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.29.mlp.gate_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.29.mlp.up_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.29.post_attention_layernorm.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.29.self_attn.W_pack.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.29.self_attn.o_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.3.input_layernorm.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.3.mlp.down_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.3.mlp.gate_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.3.mlp.up_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.3.post_attention_layernorm.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.3.self_attn.W_pack.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.3.self_attn.o_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.30.input_layernorm.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.30.mlp.down_proj.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.30.mlp.gate_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.30.mlp.up_proj.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.30.post_attention_layernorm.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.30.self_attn.W_pack.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.30.self_attn.o_proj.weight": "pytorch_model-00002-of-00003.bin",
-    "model.layers.31.input_layernorm.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.31.mlp.down_proj.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.31.mlp.gate_proj.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.31.mlp.up_proj.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.31.post_attention_layernorm.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.31.self_attn.W_pack.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.31.self_attn.o_proj.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.32.input_layernorm.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.32.mlp.down_proj.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.32.mlp.gate_proj.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.32.mlp.up_proj.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.32.post_attention_layernorm.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.32.self_attn.W_pack.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.32.self_attn.o_proj.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.33.input_layernorm.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.33.mlp.down_proj.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.33.mlp.gate_proj.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.33.mlp.up_proj.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.33.post_attention_layernorm.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.33.self_attn.W_pack.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.33.self_attn.o_proj.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.34.input_layernorm.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.34.mlp.down_proj.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.34.mlp.gate_proj.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.34.mlp.up_proj.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.34.post_attention_layernorm.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.34.self_attn.W_pack.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.34.self_attn.o_proj.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.35.input_layernorm.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.35.mlp.down_proj.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.35.mlp.gate_proj.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.35.mlp.up_proj.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.35.post_attention_layernorm.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.35.self_attn.W_pack.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.35.self_attn.o_proj.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.36.input_layernorm.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.36.mlp.down_proj.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.36.mlp.gate_proj.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.36.mlp.up_proj.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.36.post_attention_layernorm.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.36.self_attn.W_pack.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.36.self_attn.o_proj.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.37.input_layernorm.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.37.mlp.down_proj.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.37.mlp.gate_proj.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.37.mlp.up_proj.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.37.post_attention_layernorm.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.37.self_attn.W_pack.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.37.self_attn.o_proj.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.38.input_layernorm.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.38.mlp.down_proj.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.38.mlp.gate_proj.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.38.mlp.up_proj.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.38.post_attention_layernorm.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.38.self_attn.W_pack.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.38.self_attn.o_proj.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.39.input_layernorm.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.39.mlp.down_proj.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.39.mlp.gate_proj.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.39.mlp.up_proj.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.39.post_attention_layernorm.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.39.self_attn.W_pack.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.39.self_attn.o_proj.weight": "pytorch_model-00003-of-00003.bin",
-    "model.layers.4.input_layernorm.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.4.mlp.down_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.4.mlp.gate_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.4.mlp.up_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.4.post_attention_layernorm.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.4.self_attn.W_pack.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.4.self_attn.o_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.5.input_layernorm.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.5.mlp.down_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.5.mlp.gate_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.5.mlp.up_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.5.post_attention_layernorm.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.5.self_attn.W_pack.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.5.self_attn.o_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.6.input_layernorm.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.6.mlp.down_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.6.mlp.gate_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.6.mlp.up_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.6.post_attention_layernorm.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.6.self_attn.W_pack.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.6.self_attn.o_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.7.input_layernorm.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.7.mlp.down_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.7.mlp.gate_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.7.mlp.up_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.7.post_attention_layernorm.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.7.self_attn.W_pack.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.7.self_attn.o_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.8.input_layernorm.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.8.mlp.down_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.8.mlp.gate_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.8.mlp.up_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.8.post_attention_layernorm.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.8.self_attn.W_pack.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.8.self_attn.o_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.9.input_layernorm.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.9.mlp.down_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.9.mlp.gate_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.9.mlp.up_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.9.post_attention_layernorm.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.9.self_attn.W_pack.weight": "pytorch_model-00001-of-00003.bin",
-    "model.layers.9.self_attn.o_proj.weight": "pytorch_model-00001-of-00003.bin",
-    "model.norm.weight": "pytorch_model-00003-of-00003.bin"
-  }
-}

Baichuan-13B-Chat-full/quantizer.py

@@ -1,123 +0,0 @@
-# Copyright (c) 2023, Baichuan Intelligent Technology. All rights reserved.
-
-import torch
-from typing import List
-import bz2
-import base64
-import ctypes
-from transformers.utils import logging
-logger = logging.get_logger(__name__)
-
-try:
-    from cpm_kernels.kernels.base import LazyKernelCModule, KernelFunction, round_up
-
-    class Kernel:
-        def __init__(self, code: bytes, function_names: List[str]):
-            self.code = code
-            self._function_names = function_names
-            self._cmodule = LazyKernelCModule(self.code)
-
-            for name in self._function_names:
-                setattr(self, name, KernelFunction(self._cmodule, name))
-    quantization_code = "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"
-    kernels = Kernel(
-        bz2.decompress(base64.b64decode(quantization_code)),
-        [
-            "int4_to_fp16",
-            "fp16_to_int4",
-            "int8_to_fp16",
-            "fp16_to_int8",
-            "int4_to_bf16",
-            "bf16_to_int4",
-            "int8_to_bf16",
-            "bf16_to_int8",
-        ],
-    )
-except Exception as exception:
-    kernels = None
-    logger.warning("Failed to load kernels:" + str(exception))
-
-def quant4(weight: torch.Tensor, scale: torch.Tensor):
-    stream = torch.cuda.current_stream()
-    num_row = weight.size(0)
-    num_chan_fp16 = weight.size(1)
-    # 4bit
-    num_chan_int = num_chan_fp16 // 8
-    qweight = torch.zeros((num_row, num_chan_int), dtype=torch.int32, device=weight.device)
-    intweight = torch.empty(num_row, num_chan_fp16, dtype = torch.int32)
-    intweight = torch.clip(torch.round(weight.to(scale.dtype) / scale[:, None]),-16, 15).to(dtype=torch.int32)
-
-    for j in range(num_chan_int):
-        qweight[:, j] = ((intweight[:, j*8+7] & 0x0f) << 28) \
-            | ((intweight[:, j*8+6] & 0x0f) << 24) \
-            | ((intweight[:, j*8+5] & 0x0f) << 20) \
-            | ((intweight[:, j*8+4] & 0x0f) << 16) \
-            | ((intweight[:, j*8+3] & 0x0f) << 12) \
-            | ((intweight[:, j*8+2] & 0x0f) << 8) \
-            | ((intweight[:, j*8+1] & 0x0f) << 4) \
-            | ((intweight[:, j*8] & 0x0f))
-    return qweight
-
-def dequant4(qweight: torch.Tensor, scale: torch.Tensor, input: torch.Tensor):
-    stream = torch.cuda.current_stream()
-    num_row = qweight.size(0)
-    num_chan_int = qweight.size(1)
-    # 4bit
-    num_chan_fp16 = num_chan_int * 8
-    
-    out = torch.empty((num_row, num_chan_fp16), dtype=input.dtype, device=qweight.device)
-
-    blockDim = (128, 1, 1)
-    gridDim = ((num_chan_int + blockDim[0] - 1) // blockDim[0], num_row, 1)
-    if input.dtype == torch.bfloat16:
-    	kernels.int4_to_bf16(
-            gridDim,
-       	    blockDim,
-            0,
-            stream,
-            [ctypes.c_void_p(out.data_ptr()), ctypes.c_void_p(qweight.data_ptr()), 
-            ctypes.c_void_p(scale.data_ptr()), ctypes.c_int32(num_row), ctypes.c_int32(num_chan_int), ctypes.c_int32(num_chan_fp16)],
-        )
-    elif input.dtype == torch.float16:
-        kernels.int4_to_fp16(
-            gridDim,
-            blockDim,
-            0,
-            stream,
-            [ctypes.c_void_p(out.data_ptr()), ctypes.c_void_p(qweight.data_ptr()),
-            ctypes.c_void_p(scale.data_ptr()), ctypes.c_int32(num_row), ctypes.c_int32(num_chan_int), ctypes.c_int32(num_chan_fp16)],
-        )
-    return out
-
-class QLinear(torch.nn.Module):
-    def __init__(self, bits: int, weight: torch.Tensor, bias=None):
-        super().__init__()
-        self.quant_bits = bits
-        self.scale = weight.abs().max(dim=-1).values / ((2 ** (bits - 1)) - 1)
-        self.scale = self.scale.to(torch.float32)
-        if self.quant_bits == 4:
-            self.weight = quant4(weight, self.scale)
-        elif self.quant_bits == 8:
-            self.weight = torch.round(weight.to(self.scale.dtype) / self.scale[:, None]).to(torch.int8)
-        if self.quant_bits == 8:
-            self.weight = self.weight.T
-        self.bias = None
-
-    def forward(self, input):
-        if self.quant_bits == 4:
-            assert(input.dtype == torch.bfloat16 or input.dtype == torch.float16)            
-
-        if self.weight.device != input.device:
-            self.weight = self.weight.to(input.device)
-            self.scale = self.scale.to(input.device)
-        
-        if self.quant_bits == 4:
-            self.scale = self.scale.to(input.dtype)
-            rweight = dequant4(self.weight, self.scale, input).T
-            output = torch.matmul(input, rweight)
-        elif self.quant_bits == 8:
-            rweight = self.weight.to(input.dtype) * self.scale.to(input.dtype)
-            output = torch.matmul(input, rweight)
-        if self.bias is not None:
-            output = output + self.bias
-        return output

Baichuan-13B-Chat-full/train_results.json

@@ -1,7 +0,0 @@
-{
-    "epoch": 2.0,
-    "train_loss": 0.5708327819994277,
-    "train_runtime": 105327.89,
-    "train_samples_per_second": 4.797,
-    "train_steps_per_second": 0.019
-}

Baichuan-13B-Chat-full/trainer_log.jsonl

@@ -1,198 +0,0 @@
-{"current_steps": 10, "total_steps": 1974, "loss": 0.9175, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.9996834033646177e-05, "epoch": 0.01, "percentage": 0.51, "elapsed_time": "0:08:53", "remaining_time": "1 day, 5:06:00"}
-{"current_steps": 20, "total_steps": 1974, "loss": 0.7595, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.998733693645213e-05, "epoch": 0.02, "percentage": 1.01, "elapsed_time": "0:17:16", "remaining_time": "1 day, 4:07:31"}
-{"current_steps": 30, "total_steps": 1974, "loss": 0.7375, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.997151111381707e-05, "epoch": 0.03, "percentage": 1.52, "elapsed_time": "0:25:37", "remaining_time": "1 day, 3:40:03"}
-{"current_steps": 40, "total_steps": 1974, "loss": 0.7227, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.9949360574062774e-05, "epoch": 0.04, "percentage": 2.03, "elapsed_time": "0:33:53", "remaining_time": "1 day, 3:18:52"}
-{"current_steps": 50, "total_steps": 1974, "loss": 0.7147, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.9920890927418316e-05, "epoch": 0.05, "percentage": 2.53, "elapsed_time": "0:42:08", "remaining_time": "1 day, 3:01:22"}
-{"current_steps": 60, "total_steps": 1974, "loss": 0.7102, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.988610938459917e-05, "epoch": 0.06, "percentage": 3.04, "elapsed_time": "0:50:24", "remaining_time": "1 day, 2:47:46"}
-{"current_steps": 70, "total_steps": 1974, "loss": 0.7056, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.9845024754980876e-05, "epoch": 0.07, "percentage": 3.55, "elapsed_time": "0:58:37", "remaining_time": "1 day, 2:34:35"}
-{"current_steps": 80, "total_steps": 1974, "loss": 0.7128, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.979764744436784e-05, "epoch": 0.08, "percentage": 4.05, "elapsed_time": "1:06:53", "remaining_time": "1 day, 2:23:46"}
-{"current_steps": 90, "total_steps": 1974, "loss": 0.6982, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.9743989452357756e-05, "epoch": 0.09, "percentage": 4.56, "elapsed_time": "1:15:09", "remaining_time": "1 day, 2:13:11"}
-{"current_steps": 100, "total_steps": 1974, "loss": 0.7258, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.968406436930243e-05, "epoch": 0.1, "percentage": 5.07, "elapsed_time": "1:23:27", "remaining_time": "1 day, 2:04:01"}
-{"current_steps": 110, "total_steps": 1974, "loss": 0.7095, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.961788737286559e-05, "epoch": 0.11, "percentage": 5.57, "elapsed_time": "1:31:45", "remaining_time": "1 day, 1:54:50"}
-{"current_steps": 120, "total_steps": 1974, "loss": 0.7048, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.954547522417877e-05, "epoch": 0.12, "percentage": 6.08, "elapsed_time": "1:40:03", "remaining_time": "1 day, 1:45:50"}
-{"current_steps": 130, "total_steps": 1974, "loss": 0.6805, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.946684626359607e-05, "epoch": 0.13, "percentage": 6.59, "elapsed_time": "1:48:21", "remaining_time": "1 day, 1:36:59"}
-{"current_steps": 140, "total_steps": 1974, "loss": 0.6798, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.938202040604898e-05, "epoch": 0.14, "percentage": 7.09, "elapsed_time": "1:56:36", "remaining_time": "1 day, 1:27:37"}
-{"current_steps": 150, "total_steps": 1974, "loss": 0.7134, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.929101913600238e-05, "epoch": 0.15, "percentage": 7.6, "elapsed_time": "2:04:53", "remaining_time": "1 day, 1:18:36"}
-{"current_steps": 160, "total_steps": 1974, "loss": 0.6895, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.919386550201299e-05, "epoch": 0.16, "percentage": 8.11, "elapsed_time": "2:13:08", "remaining_time": "1 day, 1:09:32"}
-{"current_steps": 170, "total_steps": 1974, "loss": 0.705, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.909058411089174e-05, "epoch": 0.17, "percentage": 8.61, "elapsed_time": "2:21:25", "remaining_time": "1 day, 1:00:44"}
-{"current_steps": 180, "total_steps": 1974, "loss": 0.6712, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.8981201121471356e-05, "epoch": 0.18, "percentage": 9.12, "elapsed_time": "2:29:41", "remaining_time": "1 day, 0:51:58"}
-{"current_steps": 190, "total_steps": 1974, "loss": 0.6744, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.886574423798097e-05, "epoch": 0.19, "percentage": 9.63, "elapsed_time": "2:37:59", "remaining_time": "1 day, 0:43:26"}
-{"current_steps": 200, "total_steps": 1974, "loss": 0.6675, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.874424270302927e-05, "epoch": 0.2, "percentage": 10.13, "elapsed_time": "2:46:13", "remaining_time": "1 day, 0:34:29"}
-{"current_steps": 210, "total_steps": 1974, "loss": 0.6726, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.861672729019797e-05, "epoch": 0.21, "percentage": 10.64, "elapsed_time": "3:06:14", "remaining_time": "1 day, 2:04:28"}
-{"current_steps": 220, "total_steps": 1974, "loss": 0.6821, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.848323029624761e-05, "epoch": 0.22, "percentage": 11.14, "elapsed_time": "3:14:32", "remaining_time": "1 day, 1:50:57"}
-{"current_steps": 230, "total_steps": 1974, "loss": 0.7133, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.834378553293748e-05, "epoch": 0.23, "percentage": 11.65, "elapsed_time": "3:22:48", "remaining_time": "1 day, 1:37:50"}
-{"current_steps": 240, "total_steps": 1974, "loss": 0.6681, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.81984283184619e-05, "epoch": 0.24, "percentage": 12.16, "elapsed_time": "3:31:01", "remaining_time": "1 day, 1:24:41"}
-{"current_steps": 250, "total_steps": 1974, "loss": 0.682, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.804719546850487e-05, "epoch": 0.25, "percentage": 12.66, "elapsed_time": "3:39:18", "remaining_time": "1 day, 1:12:19"}
-{"current_steps": 260, "total_steps": 1974, "loss": 0.6755, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.789012528691558e-05, "epoch": 0.26, "percentage": 13.17, "elapsed_time": "3:47:36", "remaining_time": "1 day, 1:00:28"}
-{"current_steps": 270, "total_steps": 1974, "loss": 0.68, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.772725755600682e-05, "epoch": 0.27, "percentage": 13.68, "elapsed_time": "3:55:51", "remaining_time": "1 day, 0:48:32"}
-{"current_steps": 280, "total_steps": 1974, "loss": 0.6663, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.755863352647909e-05, "epoch": 0.28, "percentage": 14.18, "elapsed_time": "4:04:07", "remaining_time": "1 day, 0:36:58"}
-{"current_steps": 290, "total_steps": 1974, "loss": 0.6672, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.738429590697271e-05, "epoch": 0.29, "percentage": 14.69, "elapsed_time": "4:12:21", "remaining_time": "1 day, 0:25:24"}
-{"current_steps": 300, "total_steps": 1974, "loss": 0.6466, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.720428885325069e-05, "epoch": 0.3, "percentage": 15.2, "elapsed_time": "4:20:38", "remaining_time": "1 day, 0:14:20"}
-{"current_steps": 310, "total_steps": 1974, "loss": 0.6668, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.701865795701505e-05, "epoch": 0.31, "percentage": 15.7, "elapsed_time": "4:28:57", "remaining_time": "1 day, 0:03:40"}
-{"current_steps": 320, "total_steps": 1974, "loss": 0.6591, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.684682059461469e-05, "epoch": 0.32, "percentage": 16.21, "elapsed_time": "4:37:10", "remaining_time": "23:52:39"}
-{"current_steps": 330, "total_steps": 1974, "loss": 0.661, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.665063509461097e-05, "epoch": 0.33, "percentage": 16.72, "elapsed_time": "4:45:27", "remaining_time": "23:42:06"}
-{"current_steps": 340, "total_steps": 1974, "loss": 0.6749, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.644896598002736e-05, "epoch": 0.34, "percentage": 17.22, "elapsed_time": "4:53:43", "remaining_time": "23:31:35"}
-{"current_steps": 350, "total_steps": 1974, "loss": 0.6654, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.624186432907437e-05, "epoch": 0.35, "percentage": 17.73, "elapsed_time": "5:01:59", "remaining_time": "23:21:16"}
-{"current_steps": 360, "total_steps": 1974, "loss": 0.6716, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.602938259590072e-05, "epoch": 0.36, "percentage": 18.24, "elapsed_time": "5:10:16", "remaining_time": "23:11:05"}
-{"current_steps": 370, "total_steps": 1974, "loss": 0.6796, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.581157459730783e-05, "epoch": 0.37, "percentage": 18.74, "elapsed_time": "5:18:35", "remaining_time": "23:01:09"}
-{"current_steps": 380, "total_steps": 1974, "loss": 0.6794, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.558849549911931e-05, "epoch": 0.39, "percentage": 19.25, "elapsed_time": "5:26:55", "remaining_time": "22:51:20"}
-{"current_steps": 390, "total_steps": 1974, "loss": 0.6651, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.536020180220871e-05, "epoch": 0.4, "percentage": 19.76, "elapsed_time": "5:35:12", "remaining_time": "22:41:26"}
-{"current_steps": 400, "total_steps": 1974, "loss": 0.6823, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.515032675559024e-05, "epoch": 0.41, "percentage": 20.26, "elapsed_time": "5:43:27", "remaining_time": "22:31:30"}
-{"current_steps": 410, "total_steps": 1974, "loss": 0.6677, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.4912285699446786e-05, "epoch": 0.42, "percentage": 20.77, "elapsed_time": "6:04:58", "remaining_time": "23:12:15"}
-{"current_steps": 420, "total_steps": 1974, "loss": 0.6704, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.4669201313179155e-05, "epoch": 0.43, "percentage": 21.28, "elapsed_time": "6:13:18", "remaining_time": "23:01:12"}
-{"current_steps": 430, "total_steps": 1974, "loss": 0.6481, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.442113516454638e-05, "epoch": 0.44, "percentage": 21.78, "elapsed_time": "6:21:37", "remaining_time": "22:50:18"}
-{"current_steps": 440, "total_steps": 1974, "loss": 0.665, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.416815008307488e-05, "epoch": 0.45, "percentage": 22.29, "elapsed_time": "6:29:58", "remaining_time": "22:39:35"}
-{"current_steps": 450, "total_steps": 1974, "loss": 0.6658, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.391031014414514e-05, "epoch": 0.46, "percentage": 22.8, "elapsed_time": "6:38:16", "remaining_time": "22:28:49"}
-{"current_steps": 460, "total_steps": 1974, "loss": 0.6699, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.364768065276284e-05, "epoch": 0.47, "percentage": 23.3, "elapsed_time": "6:46:39", "remaining_time": "22:18:26"}
-{"current_steps": 470, "total_steps": 1974, "loss": 0.6664, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.338032812701867e-05, "epoch": 0.48, "percentage": 23.81, "elapsed_time": "6:55:02", "remaining_time": "22:08:08"}
-{"current_steps": 480, "total_steps": 1974, "loss": 0.6817, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.310832028124069e-05, "epoch": 0.49, "percentage": 24.32, "elapsed_time": "7:03:24", "remaining_time": "21:57:52"}
-{"current_steps": 490, "total_steps": 1974, "loss": 0.6791, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.283172600884393e-05, "epoch": 0.5, "percentage": 24.82, "elapsed_time": "7:11:47", "remaining_time": "21:47:43"}
-{"current_steps": 500, "total_steps": 1974, "loss": 0.6679, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.2550615364881194e-05, "epoch": 0.51, "percentage": 25.33, "elapsed_time": "7:20:10", "remaining_time": "21:37:38"}
-{"current_steps": 510, "total_steps": 1974, "loss": 0.6476, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.226505954829973e-05, "epoch": 0.52, "percentage": 25.84, "elapsed_time": "7:28:32", "remaining_time": "21:27:33"}
-{"current_steps": 520, "total_steps": 1974, "loss": 0.6721, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.197513088390813e-05, "epoch": 0.53, "percentage": 26.34, "elapsed_time": "7:36:52", "remaining_time": "21:17:30"}
-{"current_steps": 530, "total_steps": 1974, "loss": 0.6563, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.1680902804058095e-05, "epoch": 0.54, "percentage": 26.85, "elapsed_time": "7:45:10", "remaining_time": "21:07:22"}
-{"current_steps": 540, "total_steps": 1974, "loss": 0.6391, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.138244983004574e-05, "epoch": 0.55, "percentage": 27.36, "elapsed_time": "7:53:26", "remaining_time": "20:57:14"}
-{"current_steps": 550, "total_steps": 1974, "loss": 0.6672, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.107984755323697e-05, "epoch": 0.56, "percentage": 27.86, "elapsed_time": "8:01:42", "remaining_time": "20:47:11"}
-{"current_steps": 560, "total_steps": 1974, "loss": 0.6497, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.077317261592194e-05, "epoch": 0.57, "percentage": 28.37, "elapsed_time": "8:09:57", "remaining_time": "20:37:08"}
-{"current_steps": 570, "total_steps": 1974, "loss": 0.668, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.04625026919033e-05, "epoch": 0.58, "percentage": 28.88, "elapsed_time": "8:18:10", "remaining_time": "20:27:04"}
-{"current_steps": 580, "total_steps": 1974, "loss": 0.6682, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.0147916466823174e-05, "epoch": 0.59, "percentage": 29.38, "elapsed_time": "8:26:24", "remaining_time": "20:17:07"}
-{"current_steps": 590, "total_steps": 1974, "loss": 0.6352, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.982949361823388e-05, "epoch": 0.6, "percentage": 29.89, "elapsed_time": "8:34:38", "remaining_time": "20:07:14"}
-{"current_steps": 600, "total_steps": 1974, "loss": 0.6698, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.950731479541743e-05, "epoch": 0.61, "percentage": 30.4, "elapsed_time": "8:42:55", "remaining_time": "19:57:30"}
-{"current_steps": 610, "total_steps": 1974, "loss": 0.6549, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.918146159895882e-05, "epoch": 0.62, "percentage": 30.9, "elapsed_time": "9:01:46", "remaining_time": "20:11:25"}
-{"current_steps": 620, "total_steps": 1974, "loss": 0.6516, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.8852016560078605e-05, "epoch": 0.63, "percentage": 31.41, "elapsed_time": "9:10:12", "remaining_time": "20:01:35"}
-{"current_steps": 630, "total_steps": 1974, "loss": 0.6629, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.851906311972943e-05, "epoch": 0.64, "percentage": 31.91, "elapsed_time": "9:18:37", "remaining_time": "19:51:44"}
-{"current_steps": 640, "total_steps": 1974, "loss": 0.6764, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.821647502051616e-05, "epoch": 0.65, "percentage": 32.42, "elapsed_time": "9:26:58", "remaining_time": "19:41:46"}
-{"current_steps": 650, "total_steps": 1974, "loss": 0.6415, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.787708866250794e-05, "epoch": 0.66, "percentage": 32.93, "elapsed_time": "9:35:20", "remaining_time": "19:31:54"}
-{"current_steps": 660, "total_steps": 1974, "loss": 0.6463, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.7534440830144466e-05, "epoch": 0.67, "percentage": 33.43, "elapsed_time": "9:43:42", "remaining_time": "19:22:06"}
-{"current_steps": 670, "total_steps": 1974, "loss": 0.6508, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.71886183083464e-05, "epoch": 0.68, "percentage": 33.94, "elapsed_time": "9:52:04", "remaining_time": "19:12:19"}
-{"current_steps": 680, "total_steps": 1974, "loss": 0.6411, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.683970868611123e-05, "epoch": 0.69, "percentage": 34.45, "elapsed_time": "10:00:24", "remaining_time": "19:02:31"}
-{"current_steps": 690, "total_steps": 1974, "loss": 0.6266, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.648780033432891e-05, "epoch": 0.7, "percentage": 34.95, "elapsed_time": "10:08:43", "remaining_time": "18:52:44"}
-{"current_steps": 700, "total_steps": 1974, "loss": 0.6409, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.613298238339955e-05, "epoch": 0.71, "percentage": 35.46, "elapsed_time": "10:17:01", "remaining_time": "18:42:58"}
-{"current_steps": 710, "total_steps": 1974, "loss": 0.6594, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.5775344700658705e-05, "epoch": 0.72, "percentage": 35.97, "elapsed_time": "10:25:18", "remaining_time": "18:33:13"}
-{"current_steps": 720, "total_steps": 1974, "loss": 0.6427, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.5414977867616006e-05, "epoch": 0.73, "percentage": 36.47, "elapsed_time": "10:33:40", "remaining_time": "18:23:38"}
-{"current_steps": 730, "total_steps": 1974, "loss": 0.6462, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.505197315701292e-05, "epoch": 0.74, "percentage": 36.98, "elapsed_time": "10:41:59", "remaining_time": "18:14:01"}
-{"current_steps": 740, "total_steps": 1974, "loss": 0.6277, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.468642250970547e-05, "epoch": 0.75, "percentage": 37.49, "elapsed_time": "10:50:18", "remaining_time": "18:04:25"}
-{"current_steps": 750, "total_steps": 1974, "loss": 0.6551, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.431841851137764e-05, "epoch": 0.76, "percentage": 37.99, "elapsed_time": "10:58:35", "remaining_time": "17:54:48"}
-{"current_steps": 760, "total_steps": 1974, "loss": 0.6402, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.394805436909157e-05, "epoch": 0.77, "percentage": 38.5, "elapsed_time": "11:06:52", "remaining_time": "17:45:15"}
-{"current_steps": 770, "total_steps": 1974, "loss": 0.6515, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.357542388768033e-05, "epoch": 0.78, "percentage": 39.01, "elapsed_time": "11:15:12", "remaining_time": "17:35:47"}
-{"current_steps": 780, "total_steps": 1974, "loss": 0.6489, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.3200621445989226e-05, "epoch": 0.79, "percentage": 39.51, "elapsed_time": "11:23:30", "remaining_time": "17:26:18"}
-{"current_steps": 790, "total_steps": 1974, "loss": 0.6568, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.282374197297185e-05, "epoch": 0.8, "percentage": 40.02, "elapsed_time": "11:31:50", "remaining_time": "17:16:53"}
-{"current_steps": 800, "total_steps": 1974, "loss": 0.615, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.2444880923646674e-05, "epoch": 0.81, "percentage": 40.53, "elapsed_time": "11:40:07", "remaining_time": "17:07:26"}
-{"current_steps": 810, "total_steps": 1974, "loss": 0.6447, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.20641342549205e-05, "epoch": 0.82, "percentage": 41.03, "elapsed_time": "12:02:31", "remaining_time": "17:18:17"}
-{"current_steps": 820, "total_steps": 1974, "loss": 0.6159, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.168159840128472e-05, "epoch": 0.83, "percentage": 41.54, "elapsed_time": "12:10:54", "remaining_time": "17:08:37"}
-{"current_steps": 830, "total_steps": 1974, "loss": 0.6347, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.129737025039068e-05, "epoch": 0.84, "percentage": 42.05, "elapsed_time": "12:19:18", "remaining_time": "16:58:59"}
-{"current_steps": 840, "total_steps": 1974, "loss": 0.6361, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.091154711851022e-05, "epoch": 0.85, "percentage": 42.55, "elapsed_time": "12:27:42", "remaining_time": "16:49:24"}
-{"current_steps": 850, "total_steps": 1974, "loss": 0.6504, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.052422672588765e-05, "epoch": 0.86, "percentage": 43.06, "elapsed_time": "12:36:05", "remaining_time": "16:39:48"}
-{"current_steps": 860, "total_steps": 1974, "loss": 0.6467, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.013550717198948e-05, "epoch": 0.87, "percentage": 43.57, "elapsed_time": "12:44:27", "remaining_time": "16:30:14"}
-{"current_steps": 870, "total_steps": 1974, "loss": 0.6364, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.9745486910657993e-05, "epoch": 0.88, "percentage": 44.07, "elapsed_time": "12:52:51", "remaining_time": "16:20:44"}
-{"current_steps": 880, "total_steps": 1974, "loss": 0.6361, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.9354264725175185e-05, "epoch": 0.89, "percentage": 44.58, "elapsed_time": "13:01:16", "remaining_time": "16:11:15"}
-{"current_steps": 890, "total_steps": 1974, "loss": 0.6441, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.8961939703243122e-05, "epoch": 0.9, "percentage": 45.09, "elapsed_time": "13:09:39", "remaining_time": "16:01:47"}
-{"current_steps": 900, "total_steps": 1974, "loss": 0.6404, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.856861121188735e-05, "epoch": 0.91, "percentage": 45.59, "elapsed_time": "13:18:01", "remaining_time": "15:52:19"}
-{"current_steps": 910, "total_steps": 1974, "loss": 0.6307, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.8174378872289446e-05, "epoch": 0.92, "percentage": 46.1, "elapsed_time": "13:26:21", "remaining_time": "15:42:49"}
-{"current_steps": 920, "total_steps": 1974, "loss": 0.6484, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.777934253455522e-05, "epoch": 0.93, "percentage": 46.61, "elapsed_time": "13:34:42", "remaining_time": "15:33:21"}
-{"current_steps": 930, "total_steps": 1974, "loss": 0.6237, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.7383602252424985e-05, "epoch": 0.94, "percentage": 47.11, "elapsed_time": "13:43:00", "remaining_time": "15:23:53"}
-{"current_steps": 940, "total_steps": 1974, "loss": 0.6161, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.6987258257932175e-05, "epoch": 0.95, "percentage": 47.62, "elapsed_time": "13:51:20", "remaining_time": "15:14:29"}
-{"current_steps": 950, "total_steps": 1974, "loss": 0.6381, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.6590410936016895e-05, "epoch": 0.96, "percentage": 48.13, "elapsed_time": "13:59:40", "remaining_time": "15:05:05"}
-{"current_steps": 960, "total_steps": 1974, "loss": 0.6366, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.619316079910063e-05, "epoch": 0.97, "percentage": 48.63, "elapsed_time": "14:07:59", "remaining_time": "14:55:41"}
-{"current_steps": 970, "total_steps": 1974, "loss": 0.6202, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.5795608461628802e-05, "epoch": 0.98, "percentage": 49.14, "elapsed_time": "14:16:20", "remaining_time": "14:46:21"}
-{"current_steps": 980, "total_steps": 1974, "loss": 0.6334, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.5397854614587334e-05, "epoch": 0.99, "percentage": 49.65, "elapsed_time": "14:24:38", "remaining_time": "14:36:59"}
-{"current_steps": 990, "total_steps": 1974, "loss": 0.5954, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.5e-05, "epoch": 1.0, "percentage": 50.15, "elapsed_time": "14:32:58", "remaining_time": "14:27:41"}
-{"current_steps": 1000, "total_steps": 1974, "loss": 0.4963, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.460214538541267e-05, "epoch": 1.01, "percentage": 50.66, "elapsed_time": "14:41:17", "remaining_time": "14:18:23"}
-{"current_steps": 1010, "total_steps": 1974, "loss": 0.487, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.4204391538371207e-05, "epoch": 1.02, "percentage": 51.17, "elapsed_time": "15:03:28", "remaining_time": "14:22:20"}
-{"current_steps": 1020, "total_steps": 1974, "loss": 0.489, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.3806839200899377e-05, "epoch": 1.03, "percentage": 51.67, "elapsed_time": "15:11:48", "remaining_time": "14:12:48"}
-{"current_steps": 1030, "total_steps": 1974, "loss": 0.4805, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.3409589063983117e-05, "epoch": 1.04, "percentage": 52.18, "elapsed_time": "15:20:06", "remaining_time": "14:03:17"}
-{"current_steps": 1040, "total_steps": 1974, "loss": 0.4907, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.3012741742067838e-05, "epoch": 1.05, "percentage": 52.68, "elapsed_time": "15:28:27", "remaining_time": "13:53:49"}
-{"current_steps": 1050, "total_steps": 1974, "loss": 0.4719, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.261639774757503e-05, "epoch": 1.06, "percentage": 53.19, "elapsed_time": "15:36:47", "remaining_time": "13:44:22"}
-{"current_steps": 1060, "total_steps": 1974, "loss": 0.4914, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.2220657465444782e-05, "epoch": 1.07, "percentage": 53.7, "elapsed_time": "15:45:07", "remaining_time": "13:34:56"}
-{"current_steps": 1070, "total_steps": 1974, "loss": 0.4775, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.182562112771056e-05, "epoch": 1.08, "percentage": 54.2, "elapsed_time": "15:53:27", "remaining_time": "13:25:32"}
-{"current_steps": 1080, "total_steps": 1974, "loss": 0.4935, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.143138878811265e-05, "epoch": 1.09, "percentage": 54.71, "elapsed_time": "16:01:47", "remaining_time": "13:16:09"}
-{"current_steps": 1090, "total_steps": 1974, "loss": 0.5082, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.1038060296756883e-05, "epoch": 1.1, "percentage": 55.22, "elapsed_time": "16:10:09", "remaining_time": "13:06:48"}
-{"current_steps": 1100, "total_steps": 1974, "loss": 0.4868, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.064573527482482e-05, "epoch": 1.11, "percentage": 55.72, "elapsed_time": "16:18:28", "remaining_time": "12:57:26"}
-{"current_steps": 1110, "total_steps": 1974, "loss": 0.4988, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.025451308934201e-05, "epoch": 1.12, "percentage": 56.23, "elapsed_time": "16:26:44", "remaining_time": "12:48:03"}
-{"current_steps": 1120, "total_steps": 1974, "loss": 0.4653, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.9864492828010526e-05, "epoch": 1.13, "percentage": 56.74, "elapsed_time": "16:35:02", "remaining_time": "12:38:43"}
-{"current_steps": 1130, "total_steps": 1974, "loss": 0.4915, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.9475773274112354e-05, "epoch": 1.14, "percentage": 57.24, "elapsed_time": "16:43:23", "remaining_time": "12:29:25"}
-{"current_steps": 1140, "total_steps": 1974, "loss": 0.4763, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.9088452881489787e-05, "epoch": 1.16, "percentage": 57.75, "elapsed_time": "16:51:43", "remaining_time": "12:20:09"}
-{"current_steps": 1150, "total_steps": 1974, "loss": 0.4807, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.8702629749609324e-05, "epoch": 1.17, "percentage": 58.26, "elapsed_time": "17:00:05", "remaining_time": "12:10:54"}
-{"current_steps": 1160, "total_steps": 1974, "loss": 0.4653, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.8318401598715284e-05, "epoch": 1.18, "percentage": 58.76, "elapsed_time": "17:08:24", "remaining_time": "12:01:39"}
-{"current_steps": 1170, "total_steps": 1974, "loss": 0.4778, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.793586574507951e-05, "epoch": 1.19, "percentage": 59.27, "elapsed_time": "17:16:43", "remaining_time": "11:52:24"}
-{"current_steps": 1180, "total_steps": 1974, "loss": 0.4839, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.7555119076353338e-05, "epoch": 1.2, "percentage": 59.78, "elapsed_time": "17:25:04", "remaining_time": "11:43:12"}
-{"current_steps": 1190, "total_steps": 1974, "loss": 0.4718, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.7176258027028152e-05, "epoch": 1.21, "percentage": 60.28, "elapsed_time": "17:33:23", "remaining_time": "11:33:59"}
-{"current_steps": 1200, "total_steps": 1974, "loss": 0.4793, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.6799378554010773e-05, "epoch": 1.22, "percentage": 60.79, "elapsed_time": "17:41:44", "remaining_time": "11:24:49"}
-{"current_steps": 1210, "total_steps": 1974, "loss": 0.4825, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.6424576112319672e-05, "epoch": 1.23, "percentage": 61.3, "elapsed_time": "18:04:33", "remaining_time": "11:24:47"}
-{"current_steps": 1220, "total_steps": 1974, "loss": 0.4857, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.6051945630908426e-05, "epoch": 1.24, "percentage": 61.8, "elapsed_time": "18:12:53", "remaining_time": "11:15:26"}
-{"current_steps": 1230, "total_steps": 1974, "loss": 0.4802, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.5681581488622367e-05, "epoch": 1.25, "percentage": 62.31, "elapsed_time": "18:21:13", "remaining_time": "11:06:06"}
-{"current_steps": 1240, "total_steps": 1974, "loss": 0.4812, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.5313577490294538e-05, "epoch": 1.26, "percentage": 62.82, "elapsed_time": "18:29:36", "remaining_time": "10:56:49"}
-{"current_steps": 1250, "total_steps": 1974, "loss": 0.4682, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.4948026842987084e-05, "epoch": 1.27, "percentage": 63.32, "elapsed_time": "18:38:01", "remaining_time": "10:47:33"}
-{"current_steps": 1260, "total_steps": 1974, "loss": 0.4974, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.4585022132384008e-05, "epoch": 1.28, "percentage": 63.83, "elapsed_time": "18:46:25", "remaining_time": "10:38:18"}
-{"current_steps": 1270, "total_steps": 1974, "loss": 0.4737, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.4224655299341304e-05, "epoch": 1.29, "percentage": 64.34, "elapsed_time": "18:54:44", "remaining_time": "10:29:01"}
-{"current_steps": 1280, "total_steps": 1974, "loss": 0.4877, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.3867017616600456e-05, "epoch": 1.3, "percentage": 64.84, "elapsed_time": "19:03:02", "remaining_time": "10:19:44"}
-{"current_steps": 1290, "total_steps": 1974, "loss": 0.4753, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.3512199665671094e-05, "epoch": 1.31, "percentage": 65.35, "elapsed_time": "19:11:23", "remaining_time": "10:10:30"}
-{"current_steps": 1300, "total_steps": 1974, "loss": 0.4638, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.316029131388878e-05, "epoch": 1.32, "percentage": 65.86, "elapsed_time": "19:19:45", "remaining_time": "10:01:17"}
-{"current_steps": 1310, "total_steps": 1974, "loss": 0.4626, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.2811381691653607e-05, "epoch": 1.33, "percentage": 66.36, "elapsed_time": "19:28:04", "remaining_time": "9:52:03"}
-{"current_steps": 1320, "total_steps": 1974, "loss": 0.4786, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.2465559169855535e-05, "epoch": 1.34, "percentage": 66.87, "elapsed_time": "19:36:26", "remaining_time": "9:42:52"}
-{"current_steps": 1330, "total_steps": 1974, "loss": 0.4717, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.212291133749206e-05, "epoch": 1.35, "percentage": 67.38, "elapsed_time": "19:44:49", "remaining_time": "9:33:42"}
-{"current_steps": 1340, "total_steps": 1974, "loss": 0.4803, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.178352497948384e-05, "epoch": 1.36, "percentage": 67.88, "elapsed_time": "19:53:08", "remaining_time": "9:24:31"}
-{"current_steps": 1350, "total_steps": 1974, "loss": 0.4803, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.1447486054694112e-05, "epoch": 1.37, "percentage": 68.39, "elapsed_time": "20:01:28", "remaining_time": "9:15:20"}
-{"current_steps": 1360, "total_steps": 1974, "loss": 0.4739, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.1114879674157233e-05, "epoch": 1.38, "percentage": 68.9, "elapsed_time": "20:09:47", "remaining_time": "9:06:11"}
-{"current_steps": 1370, "total_steps": 1974, "loss": 0.471, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.0785790079522001e-05, "epoch": 1.39, "percentage": 69.4, "elapsed_time": "20:18:06", "remaining_time": "8:57:01"}
-{"current_steps": 1380, "total_steps": 1974, "loss": 0.4799, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.046030062171512e-05, "epoch": 1.4, "percentage": 69.91, "elapsed_time": "20:26:25", "remaining_time": "8:47:53"}
-{"current_steps": 1390, "total_steps": 1974, "loss": 0.4689, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.0138493739830352e-05, "epoch": 1.41, "percentage": 70.42, "elapsed_time": "20:34:44", "remaining_time": "8:38:46"}
-{"current_steps": 1400, "total_steps": 1974, "loss": 0.4599, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.820450940248544e-06, "epoch": 1.42, "percentage": 70.92, "elapsed_time": "20:43:03", "remaining_time": "8:29:39"}
-{"current_steps": 1410, "total_steps": 1974, "loss": 0.5019, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.506252775993882e-06, "epoch": 1.43, "percentage": 71.43, "elapsed_time": "21:01:37", "remaining_time": "8:24:39"}
-{"current_steps": 1420, "total_steps": 1974, "loss": 0.4764, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.195978826331697e-06, "epoch": 1.44, "percentage": 71.94, "elapsed_time": "21:10:01", "remaining_time": "8:15:29"}
-{"current_steps": 1430, "total_steps": 1974, "loss": 0.4579, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 8.889707676612791e-06, "epoch": 1.45, "percentage": 72.44, "elapsed_time": "21:18:24", "remaining_time": "8:06:19"}
-{"current_steps": 1440, "total_steps": 1974, "loss": 0.4592, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 8.587516898369589e-06, "epoch": 1.46, "percentage": 72.95, "elapsed_time": "21:26:47", "remaining_time": "7:57:11"}
-{"current_steps": 1450, "total_steps": 1974, "loss": 0.4861, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 8.289483029668972e-06, "epoch": 1.47, "percentage": 73.45, "elapsed_time": "21:35:09", "remaining_time": "7:48:02"}
-{"current_steps": 1460, "total_steps": 1974, "loss": 0.4753, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 7.99568155572701e-06, "epoch": 1.48, "percentage": 73.96, "elapsed_time": "21:43:33", "remaining_time": "7:38:55"}
-{"current_steps": 1470, "total_steps": 1974, "loss": 0.4929, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 7.706186889790209e-06, "epoch": 1.49, "percentage": 74.47, "elapsed_time": "21:51:52", "remaining_time": "7:29:47"}
-{"current_steps": 1480, "total_steps": 1974, "loss": 0.4594, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 7.421072354288302e-06, "epoch": 1.5, "percentage": 74.97, "elapsed_time": "22:00:14", "remaining_time": "7:20:40"}
-{"current_steps": 1490, "total_steps": 1974, "loss": 0.4912, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 7.140410162263414e-06, "epoch": 1.51, "percentage": 75.48, "elapsed_time": "22:08:35", "remaining_time": "7:11:34"}
-{"current_steps": 1500, "total_steps": 1974, "loss": 0.4681, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.86427139908008e-06, "epoch": 1.52, "percentage": 75.99, "elapsed_time": "22:16:55", "remaining_time": "7:02:28"}
-{"current_steps": 1510, "total_steps": 1974, "loss": 0.4816, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.5927260044209655e-06, "epoch": 1.53, "percentage": 76.49, "elapsed_time": "22:25:16", "remaining_time": "6:53:22"}
-{"current_steps": 1520, "total_steps": 1974, "loss": 0.4723, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.3258427545727e-06, "epoch": 1.54, "percentage": 77.0, "elapsed_time": "22:33:37", "remaining_time": "6:44:18"}
-{"current_steps": 1530, "total_steps": 1974, "loss": 0.4856, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.063689245006443e-06, "epoch": 1.55, "percentage": 77.51, "elapsed_time": "22:41:56", "remaining_time": "6:35:13"}
-{"current_steps": 1540, "total_steps": 1974, "loss": 0.4829, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.806331873257462e-06, "epoch": 1.56, "percentage": 78.01, "elapsed_time": "22:50:16", "remaining_time": "6:26:10"}
-{"current_steps": 1550, "total_steps": 1974, "loss": 0.4741, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.553835822108152e-06, "epoch": 1.57, "percentage": 78.52, "elapsed_time": "22:58:38", "remaining_time": "6:17:07"}
-{"current_steps": 1560, "total_steps": 1974, "loss": 0.4654, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.306265043078693e-06, "epoch": 1.58, "percentage": 79.03, "elapsed_time": "23:06:58", "remaining_time": "6:08:04"}
-{"current_steps": 1570, "total_steps": 1974, "loss": 0.4668, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.0636822402296165e-06, "epoch": 1.59, "percentage": 79.53, "elapsed_time": "23:15:15", "remaining_time": "5:59:02"}
-{"current_steps": 1580, "total_steps": 1974, "loss": 0.4723, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.826148854280277e-06, "epoch": 1.6, "percentage": 80.04, "elapsed_time": "23:23:36", "remaining_time": "5:50:00"}
-{"current_steps": 1590, "total_steps": 1974, "loss": 0.4639, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.593725047047293e-06, "epoch": 1.61, "percentage": 80.55, "elapsed_time": "23:31:55", "remaining_time": "5:40:59"}
-{"current_steps": 1600, "total_steps": 1974, "loss": 0.4777, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.3664696862069505e-06, "epoch": 1.62, "percentage": 81.05, "elapsed_time": "23:40:12", "remaining_time": "5:31:58"}
-{"current_steps": 1610, "total_steps": 1974, "loss": 0.4546, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.144440330385347e-06, "epoch": 1.63, "percentage": 81.56, "elapsed_time": "1 day, 0:03:58", "remaining_time": "5:26:27"}
-{"current_steps": 1620, "total_steps": 1974, "loss": 0.4543, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.927693214580075e-06, "epoch": 1.64, "percentage": 82.07, "elapsed_time": "1 day, 0:12:13", "remaining_time": "5:17:20"}
-{"current_steps": 1630, "total_steps": 1974, "loss": 0.4543, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.71628323591722e-06, "epoch": 1.65, "percentage": 82.57, "elapsed_time": "1 day, 0:20:27", "remaining_time": "5:08:13"}
-{"current_steps": 1640, "total_steps": 1974, "loss": 0.4659, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.5102639397471214e-06, "epoch": 1.66, "percentage": 83.08, "elapsed_time": "1 day, 0:28:43", "remaining_time": "4:59:07"}
-{"current_steps": 1650, "total_steps": 1974, "loss": 0.485, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.3096875060825845e-06, "epoch": 1.67, "percentage": 83.59, "elapsed_time": "1 day, 0:36:59", "remaining_time": "4:50:01"}
-{"current_steps": 1660, "total_steps": 1974, "loss": 0.4768, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.11460473638282e-06, "epoch": 1.68, "percentage": 84.09, "elapsed_time": "1 day, 0:45:17", "remaining_time": "4:40:57"}
-{"current_steps": 1670, "total_steps": 1974, "loss": 0.4635, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.925065040686642e-06, "epoch": 1.69, "percentage": 84.6, "elapsed_time": "1 day, 0:53:32", "remaining_time": "4:31:52"}
-{"current_steps": 1680, "total_steps": 1974, "loss": 0.4681, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.741116425097995e-06, "epoch": 1.7, "percentage": 85.11, "elapsed_time": "1 day, 1:01:47", "remaining_time": "4:22:48"}
-{"current_steps": 1690, "total_steps": 1974, "loss": 0.4492, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.5628054796271063e-06, "epoch": 1.71, "percentage": 85.61, "elapsed_time": "1 day, 1:10:02", "remaining_time": "4:13:45"}
-{"current_steps": 1700, "total_steps": 1974, "loss": 0.4664, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.390177366390273e-06, "epoch": 1.72, "percentage": 86.12, "elapsed_time": "1 day, 1:18:18", "remaining_time": "4:04:42"}
-{"current_steps": 1710, "total_steps": 1974, "loss": 0.48, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.22327580817136e-06, "epoch": 1.73, "percentage": 86.63, "elapsed_time": "1 day, 1:26:34", "remaining_time": "3:55:40"}
-{"current_steps": 1720, "total_steps": 1974, "loss": 0.4616, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.0621430773477947e-06, "epoch": 1.74, "percentage": 87.13, "elapsed_time": "1 day, 1:34:47", "remaining_time": "3:46:38"}
-{"current_steps": 1730, "total_steps": 1974, "loss": 0.4854, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.906819985183908e-06, "epoch": 1.75, "percentage": 87.64, "elapsed_time": "1 day, 1:43:03", "remaining_time": "3:37:38"}
-{"current_steps": 1740, "total_steps": 1974, "loss": 0.4846, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.7573458714944063e-06, "epoch": 1.76, "percentage": 88.15, "elapsed_time": "1 day, 1:51:18", "remaining_time": "3:28:37"}
-{"current_steps": 1750, "total_steps": 1974, "loss": 0.4552, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.6137585946804674e-06, "epoch": 1.77, "percentage": 88.65, "elapsed_time": "1 day, 1:59:35", "remaining_time": "3:19:37"}
-{"current_steps": 1760, "total_steps": 1974, "loss": 0.4615, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.4760945221410638e-06, "epoch": 1.78, "percentage": 89.16, "elapsed_time": "1 day, 2:07:51", "remaining_time": "3:10:38"}
-{"current_steps": 1770, "total_steps": 1974, "loss": 0.4735, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.3443885210619428e-06, "epoch": 1.79, "percentage": 89.67, "elapsed_time": "1 day, 2:16:10", "remaining_time": "3:01:39"}
-{"current_steps": 1780, "total_steps": 1974, "loss": 0.4705, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.2186739495845477e-06, "epoch": 1.8, "percentage": 90.17, "elapsed_time": "1 day, 2:24:26", "remaining_time": "2:52:41"}
-{"current_steps": 1790, "total_steps": 1974, "loss": 0.4653, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.0989826483571552e-06, "epoch": 1.81, "percentage": 90.68, "elapsed_time": "1 day, 2:32:40", "remaining_time": "2:43:42"}
-{"current_steps": 1800, "total_steps": 1974, "loss": 0.4632, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.85344932470364e-07, "epoch": 1.82, "percentage": 91.19, "elapsed_time": "1 day, 2:40:51", "remaining_time": "2:34:44"}
-{"current_steps": 1810, "total_steps": 1974, "loss": 0.481, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 8.77789583778979e-07, "epoch": 1.83, "percentage": 91.69, "elapsed_time": "1 day, 2:59:37", "remaining_time": "2:26:45"}
-{"current_steps": 1820, "total_steps": 1974, "loss": 0.4773, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 7.763438436122122e-07, "epoch": 1.84, "percentage": 92.2, "elapsed_time": "1 day, 3:07:54", "remaining_time": "2:17:44"}
-{"current_steps": 1830, "total_steps": 1974, "loss": 0.4791, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.810334058740736e-07, "epoch": 1.85, "percentage": 92.71, "elapsed_time": "1 day, 3:16:15", "remaining_time": "2:08:45"}
-{"current_steps": 1840, "total_steps": 1974, "loss": 0.4793, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.918824105356797e-07, "epoch": 1.86, "percentage": 93.21, "elapsed_time": "1 day, 3:24:34", "remaining_time": "1:59:46"}
-{"current_steps": 1850, "total_steps": 1974, "loss": 0.4647, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.08913437521169e-07, "epoch": 1.87, "percentage": 93.72, "elapsed_time": "1 day, 3:32:49", "remaining_time": "1:50:47"}
-{"current_steps": 1860, "total_steps": 1974, "loss": 0.4765, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.3214750098869995e-07, "epoch": 1.88, "percentage": 94.22, "elapsed_time": "1 day, 3:41:06", "remaining_time": "1:41:48"}
-{"current_steps": 1870, "total_steps": 1974, "loss": 0.4787, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.616040440080432e-07, "epoch": 1.89, "percentage": 94.73, "elapsed_time": "1 day, 3:49:26", "remaining_time": "1:32:50"}
-{"current_steps": 1880, "total_steps": 1974, "loss": 0.4723, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.973009336361021e-07, "epoch": 1.9, "percentage": 95.24, "elapsed_time": "1 day, 3:57:44", "remaining_time": "1:23:53"}
-{"current_steps": 1890, "total_steps": 1974, "loss": 0.453, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.392544563915883e-07, "epoch": 1.91, "percentage": 95.74, "elapsed_time": "1 day, 4:06:01", "remaining_time": "1:14:56"}
-{"current_steps": 1900, "total_steps": 1974, "loss": 0.4561, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.8747931413001795e-07, "epoch": 1.93, "percentage": 96.25, "elapsed_time": "1 day, 4:14:18", "remaining_time": "1:05:59"}
-{"current_steps": 1910, "total_steps": 1974, "loss": 0.4612, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.4198862032005488e-07, "epoch": 1.94, "percentage": 96.76, "elapsed_time": "1 day, 4:22:35", "remaining_time": "0:57:03"}
-{"current_steps": 1920, "total_steps": 1974, "loss": 0.4774, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.0279389672218365e-07, "epoch": 1.95, "percentage": 97.26, "elapsed_time": "1 day, 4:30:53", "remaining_time": "0:48:07"}
-{"current_steps": 1930, "total_steps": 1974, "loss": 0.4635, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.990507047049676e-08, "epoch": 1.96, "percentage": 97.77, "elapsed_time": "1 day, 4:39:09", "remaining_time": "0:39:11"}
-{"current_steps": 1940, "total_steps": 1974, "loss": 0.4761, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.3330471558378213e-08, "epoch": 1.97, "percentage": 98.28, "elapsed_time": "1 day, 4:47:22", "remaining_time": "0:30:16"}
-{"current_steps": 1950, "total_steps": 1974, "loss": 0.4593, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.3076830728713252e-08, "epoch": 1.98, "percentage": 98.78, "elapsed_time": "1 day, 4:55:39", "remaining_time": "0:21:21"}
-{"current_steps": 1960, "total_steps": 1974, "loss": 0.4736, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.149277769132658e-09, "epoch": 1.99, "percentage": 99.29, "elapsed_time": "1 day, 5:03:51", "remaining_time": "0:12:27"}
-{"current_steps": 1970, "total_steps": 1974, "loss": 0.4486, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.551340212760377e-09, "epoch": 2.0, "percentage": 99.8, "elapsed_time": "1 day, 5:12:08", "remaining_time": "0:03:33"}
-{"current_steps": 1974, "total_steps": 1974, "loss": null, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 2.0, "percentage": 100.0, "elapsed_time": "1 day, 5:15:27", "remaining_time": "0:00:00"}

Baichuan-13B-Chat-lora-Task/README.md

@@ -0,0 +1,9 @@
+---
+library_name: peft
+---
+## Training procedure
+
+### Framework versions
+
+
+- PEFT 0.5.0

Baichuan-13B-Chat-lora-Task/adapter_config.json

@@ -0,0 +1,20 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": "baichuan-inc/Baichuan-13B-Chat",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "lora_alpha": 32.0,
+  "lora_dropout": 0.1,
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 8,
+  "revision": null,
+  "target_modules": [
+    "W_pack"
+  ],
+  "task_type": "CAUSAL_LM"
+}

Baichuan-13B-Chat-full/pytorch_model-00001-of-00003.bin → Baichuan-13B-Chat-lora-Task/adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8ba122d5818b6681b04be1cf1cbd5a8d8bba0b187481b1481fccdc50e3efe479
-size 9972280267
+oid sha256:84594037c53e9a300bbdebbc3534f704f22f429b6b912fe36e32aeec5a928e3a
+size 26243422

Baichuan-13B-Chat-lora-Task/all_results.json

@@ -0,0 +1,11 @@
+{
+    "epoch": 2.0,
+    "eval_loss": 0.41885045170783997,
+    "eval_runtime": 69.6555,
+    "eval_samples_per_second": 15.821,
+    "eval_steps_per_second": 1.594,
+    "train_loss": 0.46561298847898175,
+    "train_runtime": 35477.4141,
+    "train_samples_per_second": 6.149,
+    "train_steps_per_second": 0.077
+}

Baichuan-13B-Chat-lora-Task/eval_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 2.0,
+    "eval_loss": 0.41885045170783997,
+    "eval_runtime": 69.6555,
+    "eval_samples_per_second": 15.821,
+    "eval_steps_per_second": 1.594
+}

Baichuan-13B-Chat-lora-Task/train_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 2.0,
+    "train_loss": 0.46561298847898175,
+    "train_runtime": 35477.4141,
+    "train_samples_per_second": 6.149,
+    "train_steps_per_second": 0.077
+}

Baichuan-13B-Chat-lora-Task/trainer_log.jsonl

@@ -0,0 +1,287 @@
+{"current_steps": 10, "total_steps": 2726, "loss": 1.5452, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.999865525734509e-05, "epoch": 0.01, "percentage": 0.37, "elapsed_time": "0:02:32", "remaining_time": "11:29:51"}
+{"current_steps": 20, "total_steps": 2726, "loss": 1.3291, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.9994621174046976e-05, "epoch": 0.01, "percentage": 0.73, "elapsed_time": "0:04:36", "remaining_time": "10:23:36"}
+{"current_steps": 30, "total_steps": 2726, "loss": 1.1048, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.9986985231938546e-05, "epoch": 0.02, "percentage": 1.1, "elapsed_time": "0:06:36", "remaining_time": "9:54:30"}
+{"current_steps": 40, "total_steps": 2726, "loss": 0.9155, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.99760306731191e-05, "epoch": 0.03, "percentage": 1.47, "elapsed_time": "0:08:47", "remaining_time": "9:50:47"}
+{"current_steps": 50, "total_steps": 2726, "loss": 0.8856, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.9961758952505326e-05, "epoch": 0.04, "percentage": 1.83, "elapsed_time": "0:10:46", "remaining_time": "9:36:34"}
+{"current_steps": 60, "total_steps": 2726, "loss": 0.8052, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.9944171965578836e-05, "epoch": 0.04, "percentage": 2.2, "elapsed_time": "0:13:03", "remaining_time": "9:40:26"}
+{"current_steps": 70, "total_steps": 2726, "loss": 0.7332, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.992327204813435e-05, "epoch": 0.05, "percentage": 2.57, "elapsed_time": "0:15:01", "remaining_time": "9:30:01"}
+{"current_steps": 80, "total_steps": 2726, "loss": 0.6982, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.989906197596955e-05, "epoch": 0.06, "percentage": 2.93, "elapsed_time": "0:17:01", "remaining_time": "9:22:57"}
+{"current_steps": 90, "total_steps": 2726, "loss": 0.6811, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.987154496451635e-05, "epoch": 0.07, "percentage": 3.3, "elapsed_time": "0:18:51", "remaining_time": "9:12:22"}
+{"current_steps": 100, "total_steps": 2726, "loss": 0.6323, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.984072466841389e-05, "epoch": 0.07, "percentage": 3.67, "elapsed_time": "0:21:04", "remaining_time": "9:13:18"}
+{"current_steps": 110, "total_steps": 2726, "loss": 0.6234, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.981016546765289e-05, "epoch": 0.08, "percentage": 4.04, "elapsed_time": "0:23:02", "remaining_time": "9:07:57"}
+{"current_steps": 120, "total_steps": 2726, "loss": 0.6125, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.977308057009269e-05, "epoch": 0.09, "percentage": 4.4, "elapsed_time": "0:25:11", "remaining_time": "9:07:04"}
+{"current_steps": 130, "total_steps": 2726, "loss": 0.5977, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.97327054653146e-05, "epoch": 0.1, "percentage": 4.77, "elapsed_time": "0:27:29", "remaining_time": "9:08:54"}
+{"current_steps": 140, "total_steps": 2726, "loss": 0.6179, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.968904551569013e-05, "epoch": 0.1, "percentage": 5.14, "elapsed_time": "0:29:47", "remaining_time": "9:10:22"}
+{"current_steps": 150, "total_steps": 2726, "loss": 0.5879, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.9642106519863544e-05, "epoch": 0.11, "percentage": 5.5, "elapsed_time": "0:31:46", "remaining_time": "9:05:40"}
+{"current_steps": 160, "total_steps": 2726, "loss": 0.5472, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.959189471198171e-05, "epoch": 0.12, "percentage": 5.87, "elapsed_time": "0:33:38", "remaining_time": "8:59:28"}
+{"current_steps": 170, "total_steps": 2726, "loss": 0.602, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.953841676086613e-05, "epoch": 0.12, "percentage": 6.24, "elapsed_time": "0:35:33", "remaining_time": "8:54:43"}
+{"current_steps": 180, "total_steps": 2726, "loss": 0.5773, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.9481679769127275e-05, "epoch": 0.13, "percentage": 6.6, "elapsed_time": "0:37:48", "remaining_time": "8:54:48"}
+{"current_steps": 190, "total_steps": 2726, "loss": 0.5615, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.9421691272221167e-05, "epoch": 0.14, "percentage": 6.97, "elapsed_time": "0:39:45", "remaining_time": "8:50:36"}
+{"current_steps": 200, "total_steps": 2726, "loss": 0.5704, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.935845923744865e-05, "epoch": 0.15, "percentage": 7.34, "elapsed_time": "0:41:50", "remaining_time": "8:48:29"}
+{"current_steps": 200, "total_steps": 2726, "loss": null, "eval_loss": 0.5564362406730652, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 0.15, "percentage": 7.34, "elapsed_time": "0:41:50", "remaining_time": "8:48:29"}
+{"current_steps": 210, "total_steps": 2726, "loss": 0.5803, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.9291992062897183e-05, "epoch": 0.15, "percentage": 7.7, "elapsed_time": "0:44:59", "remaining_time": "8:59:06"}
+{"current_steps": 220, "total_steps": 2726, "loss": 0.5655, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.922229857632545e-05, "epoch": 0.16, "percentage": 8.07, "elapsed_time": "0:46:42", "remaining_time": "8:52:03"}
+{"current_steps": 230, "total_steps": 2726, "loss": 0.5769, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.9149388033990966e-05, "epoch": 0.17, "percentage": 8.44, "elapsed_time": "0:49:02", "remaining_time": "8:52:13"}
+{"current_steps": 240, "total_steps": 2726, "loss": 0.5862, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.9073270119420635e-05, "epoch": 0.18, "percentage": 8.8, "elapsed_time": "0:51:04", "remaining_time": "8:49:08"}
+{"current_steps": 250, "total_steps": 2726, "loss": 0.5506, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.899395494212471e-05, "epoch": 0.18, "percentage": 9.17, "elapsed_time": "0:53:13", "remaining_time": "8:47:07"}
+{"current_steps": 260, "total_steps": 2726, "loss": 0.5183, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.891145303625408e-05, "epoch": 0.19, "percentage": 9.54, "elapsed_time": "0:55:40", "remaining_time": "8:48:05"}
+{"current_steps": 270, "total_steps": 2726, "loss": 0.575, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.882577535920121e-05, "epoch": 0.2, "percentage": 9.9, "elapsed_time": "0:58:06", "remaining_time": "8:48:33"}
+{"current_steps": 280, "total_steps": 2726, "loss": 0.5359, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.8736933290144815e-05, "epoch": 0.21, "percentage": 10.27, "elapsed_time": "1:00:24", "remaining_time": "8:47:41"}
+{"current_steps": 290, "total_steps": 2726, "loss": 0.5302, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.8644938628538606e-05, "epoch": 0.21, "percentage": 10.64, "elapsed_time": "1:02:22", "remaining_time": "8:43:57"}
+{"current_steps": 300, "total_steps": 2726, "loss": 0.5399, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.8549803592544076e-05, "epoch": 0.22, "percentage": 11.01, "elapsed_time": "1:04:13", "remaining_time": "8:39:25"}
+{"current_steps": 310, "total_steps": 2726, "loss": 0.5379, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.845154081740783e-05, "epoch": 0.23, "percentage": 11.37, "elapsed_time": "1:06:13", "remaining_time": "8:36:11"}
+{"current_steps": 320, "total_steps": 2726, "loss": 0.5299, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.835016335378343e-05, "epoch": 0.23, "percentage": 11.74, "elapsed_time": "1:08:08", "remaining_time": "8:32:17"}
+{"current_steps": 330, "total_steps": 2726, "loss": 0.535, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.8245684665998073e-05, "epoch": 0.24, "percentage": 12.11, "elapsed_time": "1:10:23", "remaining_time": "8:31:06"}
+{"current_steps": 340, "total_steps": 2726, "loss": 0.5145, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.813811863026436e-05, "epoch": 0.25, "percentage": 12.47, "elapsed_time": "1:12:33", "remaining_time": "8:29:13"}
+{"current_steps": 350, "total_steps": 2726, "loss": 0.5126, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.802747953283732e-05, "epoch": 0.26, "percentage": 12.84, "elapsed_time": "1:14:40", "remaining_time": "8:26:55"}
+{"current_steps": 360, "total_steps": 2726, "loss": 0.5012, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.791378206811704e-05, "epoch": 0.26, "percentage": 13.21, "elapsed_time": "1:16:47", "remaining_time": "8:24:39"}
+{"current_steps": 370, "total_steps": 2726, "loss": 0.5373, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.7797041336696995e-05, "epoch": 0.27, "percentage": 13.57, "elapsed_time": "1:19:07", "remaining_time": "8:23:47"}
+{"current_steps": 380, "total_steps": 2726, "loss": 0.4778, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.767727284335852e-05, "epoch": 0.28, "percentage": 13.94, "elapsed_time": "1:21:18", "remaining_time": "8:21:59"}
+{"current_steps": 390, "total_steps": 2726, "loss": 0.4894, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.755449249501155e-05, "epoch": 0.29, "percentage": 14.31, "elapsed_time": "1:23:03", "remaining_time": "8:17:32"}
+{"current_steps": 400, "total_steps": 2726, "loss": 0.4986, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.7428716598581934e-05, "epoch": 0.29, "percentage": 14.67, "elapsed_time": "1:25:10", "remaining_time": "8:15:20"}
+{"current_steps": 400, "total_steps": 2726, "loss": null, "eval_loss": 0.4967592656612396, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 0.29, "percentage": 14.67, "elapsed_time": "1:25:10", "remaining_time": "8:15:20"}
+{"current_steps": 410, "total_steps": 2726, "loss": 0.4928, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.729996185884571e-05, "epoch": 0.3, "percentage": 15.04, "elapsed_time": "1:28:14", "remaining_time": "8:18:25"}
+{"current_steps": 420, "total_steps": 2726, "loss": 0.4807, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.716824537621042e-05, "epoch": 0.31, "percentage": 15.41, "elapsed_time": "1:30:05", "remaining_time": "8:14:40"}
+{"current_steps": 430, "total_steps": 2726, "loss": 0.4734, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.703358464444397e-05, "epoch": 0.32, "percentage": 15.77, "elapsed_time": "1:32:27", "remaining_time": "8:13:42"}
+{"current_steps": 440, "total_steps": 2726, "loss": 0.4603, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.689599754835122e-05, "epoch": 0.32, "percentage": 16.14, "elapsed_time": "1:34:29", "remaining_time": "8:10:56"}
+{"current_steps": 450, "total_steps": 2726, "loss": 0.4661, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.6755502361398616e-05, "epoch": 0.33, "percentage": 16.51, "elapsed_time": "1:36:31", "remaining_time": "8:08:14"}
+{"current_steps": 460, "total_steps": 2726, "loss": 0.5072, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.6612117743287234e-05, "epoch": 0.34, "percentage": 16.87, "elapsed_time": "1:38:36", "remaining_time": "8:05:42"}
+{"current_steps": 470, "total_steps": 2726, "loss": 0.5056, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.646586273747452e-05, "epoch": 0.34, "percentage": 17.24, "elapsed_time": "1:40:39", "remaining_time": "8:03:07"}
+{"current_steps": 480, "total_steps": 2726, "loss": 0.4391, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.631675676864503e-05, "epoch": 0.35, "percentage": 17.61, "elapsed_time": "1:43:01", "remaining_time": "8:02:03"}
+{"current_steps": 490, "total_steps": 2726, "loss": 0.483, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.6164819640130595e-05, "epoch": 0.36, "percentage": 17.98, "elapsed_time": "1:45:28", "remaining_time": "8:01:17"}
+{"current_steps": 500, "total_steps": 2726, "loss": 0.4774, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.601007153128014e-05, "epoch": 0.37, "percentage": 18.34, "elapsed_time": "1:47:35", "remaining_time": "7:59:00"}
+{"current_steps": 510, "total_steps": 2726, "loss": 0.4848, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.5852532994779606e-05, "epoch": 0.37, "percentage": 18.71, "elapsed_time": "1:49:45", "remaining_time": "7:56:55"}
+{"current_steps": 520, "total_steps": 2726, "loss": 0.5099, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.5692224953922266e-05, "epoch": 0.38, "percentage": 19.08, "elapsed_time": "1:51:38", "remaining_time": "7:53:35"}
+{"current_steps": 530, "total_steps": 2726, "loss": 0.4851, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.5529168699829805e-05, "epoch": 0.39, "percentage": 19.44, "elapsed_time": "1:53:42", "remaining_time": "7:51:09"}
+{"current_steps": 540, "total_steps": 2726, "loss": 0.498, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.536338588862459e-05, "epoch": 0.4, "percentage": 19.81, "elapsed_time": "1:55:51", "remaining_time": "7:48:59"}
+{"current_steps": 550, "total_steps": 2726, "loss": 0.5398, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.519489853855341e-05, "epoch": 0.4, "percentage": 20.18, "elapsed_time": "1:58:05", "remaining_time": "7:47:11"}
+{"current_steps": 560, "total_steps": 2726, "loss": 0.4566, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.50237290270632e-05, "epoch": 0.41, "percentage": 20.54, "elapsed_time": "1:59:55", "remaining_time": "7:43:50"}
+{"current_steps": 570, "total_steps": 2726, "loss": 0.4942, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.484990008782893e-05, "epoch": 0.42, "percentage": 20.91, "elapsed_time": "2:01:51", "remaining_time": "7:40:56"}
+{"current_steps": 580, "total_steps": 2726, "loss": 0.4625, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.467343480773433e-05, "epoch": 0.43, "percentage": 21.28, "elapsed_time": "2:04:02", "remaining_time": "7:38:55"}
+{"current_steps": 590, "total_steps": 2726, "loss": 0.4703, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.44943566238056e-05, "epoch": 0.43, "percentage": 21.64, "elapsed_time": "2:06:02", "remaining_time": "7:36:17"}
+{"current_steps": 600, "total_steps": 2726, "loss": 0.464, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.431268932009865e-05, "epoch": 0.44, "percentage": 22.01, "elapsed_time": "2:08:29", "remaining_time": "7:35:16"}
+{"current_steps": 600, "total_steps": 2726, "loss": null, "eval_loss": 0.46821752190589905, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 0.44, "percentage": 22.01, "elapsed_time": "2:08:29", "remaining_time": "7:35:16"}
+{"current_steps": 610, "total_steps": 2726, "loss": 0.4739, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.412845702454024e-05, "epoch": 0.45, "percentage": 22.38, "elapsed_time": "2:11:46", "remaining_time": "7:37:07"}
+{"current_steps": 620, "total_steps": 2726, "loss": 0.4529, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.394168420572349e-05, "epoch": 0.45, "percentage": 22.74, "elapsed_time": "2:14:01", "remaining_time": "7:35:15"}
+{"current_steps": 630, "total_steps": 2726, "loss": 0.4433, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.3752395669658086e-05, "epoch": 0.46, "percentage": 23.11, "elapsed_time": "2:16:41", "remaining_time": "7:34:47"}
+{"current_steps": 640, "total_steps": 2726, "loss": 0.459, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.356061655647571e-05, "epoch": 0.47, "percentage": 23.48, "elapsed_time": "2:18:53", "remaining_time": "7:32:41"}
+{"current_steps": 650, "total_steps": 2726, "loss": 0.4877, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.336637233709107e-05, "epoch": 0.48, "percentage": 23.84, "elapsed_time": "2:20:42", "remaining_time": "7:29:22"}
+{"current_steps": 660, "total_steps": 2726, "loss": 0.4676, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.316968880981903e-05, "epoch": 0.48, "percentage": 24.21, "elapsed_time": "2:22:37", "remaining_time": "7:26:26"}
+{"current_steps": 670, "total_steps": 2726, "loss": 0.4782, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.2970592096948236e-05, "epoch": 0.49, "percentage": 24.58, "elapsed_time": "2:24:37", "remaining_time": "7:23:48"}
+{"current_steps": 680, "total_steps": 2726, "loss": 0.4778, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.276910864127168e-05, "epoch": 0.5, "percentage": 24.94, "elapsed_time": "2:26:47", "remaining_time": "7:21:39"}
+{"current_steps": 690, "total_steps": 2726, "loss": 0.454, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.25652652025748e-05, "epoch": 0.51, "percentage": 25.31, "elapsed_time": "2:28:42", "remaining_time": "7:18:47"}
+{"current_steps": 700, "total_steps": 2726, "loss": 0.4399, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.235908885408133e-05, "epoch": 0.51, "percentage": 25.68, "elapsed_time": "2:30:54", "remaining_time": "7:16:47"}
+{"current_steps": 710, "total_steps": 2726, "loss": 0.5106, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.215060697885767e-05, "epoch": 0.52, "percentage": 26.05, "elapsed_time": "2:32:57", "remaining_time": "7:14:19"}
+{"current_steps": 720, "total_steps": 2726, "loss": 0.4683, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.1939847266176e-05, "epoch": 0.53, "percentage": 26.41, "elapsed_time": "2:35:12", "remaining_time": "7:12:26"}
+{"current_steps": 730, "total_steps": 2726, "loss": 0.4896, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.172683770783677e-05, "epoch": 0.54, "percentage": 26.78, "elapsed_time": "2:37:07", "remaining_time": "7:09:37"}
+{"current_steps": 740, "total_steps": 2726, "loss": 0.4633, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.1511606594451016e-05, "epoch": 0.54, "percentage": 27.15, "elapsed_time": "2:39:14", "remaining_time": "7:07:20"}
+{"current_steps": 750, "total_steps": 2726, "loss": 0.4486, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.1294182511682946e-05, "epoch": 0.55, "percentage": 27.51, "elapsed_time": "2:41:16", "remaining_time": "7:04:53"}
+{"current_steps": 760, "total_steps": 2726, "loss": 0.4706, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.1074594336453384e-05, "epoch": 0.56, "percentage": 27.88, "elapsed_time": "2:43:11", "remaining_time": "7:02:09"}
+{"current_steps": 770, "total_steps": 2726, "loss": 0.4797, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.085287123310455e-05, "epoch": 0.56, "percentage": 28.25, "elapsed_time": "2:44:59", "remaining_time": "6:59:07"}
+{"current_steps": 780, "total_steps": 2726, "loss": 0.4251, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.062904264952657e-05, "epoch": 0.57, "percentage": 28.61, "elapsed_time": "2:47:04", "remaining_time": "6:56:49"}
+{"current_steps": 790, "total_steps": 2726, "loss": 0.4687, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.0403138313246435e-05, "epoch": 0.58, "percentage": 28.98, "elapsed_time": "2:49:15", "remaining_time": "6:54:47"}
+{"current_steps": 800, "total_steps": 2726, "loss": 0.48, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.017518822747976e-05, "epoch": 0.59, "percentage": 29.35, "elapsed_time": "2:51:34", "remaining_time": "6:53:03"}
+{"current_steps": 800, "total_steps": 2726, "loss": null, "eval_loss": 0.4554018974304199, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 0.59, "percentage": 29.35, "elapsed_time": "2:51:34", "remaining_time": "6:53:03"}
+{"current_steps": 810, "total_steps": 2726, "loss": 0.4662, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.994522266714594e-05, "epoch": 0.59, "percentage": 29.71, "elapsed_time": "2:54:51", "remaining_time": "6:53:36"}
+{"current_steps": 820, "total_steps": 2726, "loss": 0.4837, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.9713272174847246e-05, "epoch": 0.6, "percentage": 30.08, "elapsed_time": "2:57:25", "remaining_time": "6:52:24"}
+{"current_steps": 830, "total_steps": 2726, "loss": 0.5014, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.947936755681229e-05, "epoch": 0.61, "percentage": 30.45, "elapsed_time": "2:59:33", "remaining_time": "6:50:10"}
+{"current_steps": 840, "total_steps": 2726, "loss": 0.4764, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.92435398788046e-05, "epoch": 0.62, "percentage": 30.81, "elapsed_time": "3:01:44", "remaining_time": "6:48:02"}
+{"current_steps": 850, "total_steps": 2726, "loss": 0.4587, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.9005820461996604e-05, "epoch": 0.62, "percentage": 31.18, "elapsed_time": "3:03:52", "remaining_time": "6:45:49"}
+{"current_steps": 860, "total_steps": 2726, "loss": 0.484, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.876624087880979e-05, "epoch": 0.63, "percentage": 31.55, "elapsed_time": "3:06:00", "remaining_time": "6:43:35"}
+{"current_steps": 870, "total_steps": 2726, "loss": 0.4285, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.85248329487214e-05, "epoch": 0.64, "percentage": 31.91, "elapsed_time": "3:08:03", "remaining_time": "6:41:10"}
+{"current_steps": 880, "total_steps": 2726, "loss": 0.4742, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.828162873403843e-05, "epoch": 0.65, "percentage": 32.28, "elapsed_time": "3:10:09", "remaining_time": "6:38:53"}
+{"current_steps": 890, "total_steps": 2726, "loss": 0.4533, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.803666053563926e-05, "epoch": 0.65, "percentage": 32.65, "elapsed_time": "3:12:20", "remaining_time": "6:36:48"}
+{"current_steps": 900, "total_steps": 2726, "loss": 0.4395, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.778996088868365e-05, "epoch": 0.66, "percentage": 33.02, "elapsed_time": "3:14:34", "remaining_time": "6:34:47"}
+{"current_steps": 910, "total_steps": 2726, "loss": 0.4519, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.754156255829167e-05, "epoch": 0.67, "percentage": 33.38, "elapsed_time": "3:16:42", "remaining_time": "6:32:33"}
+{"current_steps": 920, "total_steps": 2726, "loss": 0.4684, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.7291498535191996e-05, "epoch": 0.67, "percentage": 33.75, "elapsed_time": "3:18:41", "remaining_time": "6:30:02"}
+{"current_steps": 930, "total_steps": 2726, "loss": 0.4259, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.703980203134029e-05, "epoch": 0.68, "percentage": 34.12, "elapsed_time": "3:20:24", "remaining_time": "6:27:00"}
+{"current_steps": 940, "total_steps": 2726, "loss": 0.4576, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.678650647550822e-05, "epoch": 0.69, "percentage": 34.48, "elapsed_time": "3:22:27", "remaining_time": "6:24:40"}
+{"current_steps": 950, "total_steps": 2726, "loss": 0.446, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.6531645508843636e-05, "epoch": 0.7, "percentage": 34.85, "elapsed_time": "3:24:21", "remaining_time": "6:22:01"}
+{"current_steps": 960, "total_steps": 2726, "loss": 0.4439, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.6275252980402544e-05, "epoch": 0.7, "percentage": 35.22, "elapsed_time": "3:26:18", "remaining_time": "6:19:31"}
+{"current_steps": 970, "total_steps": 2726, "loss": 0.476, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.601736294265354e-05, "epoch": 0.71, "percentage": 35.58, "elapsed_time": "3:28:32", "remaining_time": "6:17:31"}
+{"current_steps": 980, "total_steps": 2726, "loss": 0.4153, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.5758009646955115e-05, "epoch": 0.72, "percentage": 35.95, "elapsed_time": "3:30:27", "remaining_time": "6:14:57"}
+{"current_steps": 990, "total_steps": 2726, "loss": 0.4444, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.5497227539006614e-05, "epoch": 0.73, "percentage": 36.32, "elapsed_time": "3:32:34", "remaining_time": "6:12:46"}
+{"current_steps": 1000, "total_steps": 2726, "loss": 0.4395, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.523505125427341e-05, "epoch": 0.73, "percentage": 36.68, "elapsed_time": "3:34:52", "remaining_time": "6:10:52"}
+{"current_steps": 1000, "total_steps": 2726, "loss": null, "eval_loss": 0.44357746839523315, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 0.73, "percentage": 36.68, "elapsed_time": "3:34:52", "remaining_time": "6:10:52"}
+{"current_steps": 1010, "total_steps": 2726, "loss": 0.4387, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.497151561338678e-05, "epoch": 0.74, "percentage": 37.05, "elapsed_time": "3:38:14", "remaining_time": "6:10:46"}
+{"current_steps": 1020, "total_steps": 2726, "loss": 0.4559, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.470665561751928e-05, "epoch": 0.75, "percentage": 37.42, "elapsed_time": "3:40:04", "remaining_time": "6:08:04"}
+{"current_steps": 1030, "total_steps": 2726, "loss": 0.4349, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.444050644373611e-05, "epoch": 0.76, "percentage": 37.78, "elapsed_time": "3:42:11", "remaining_time": "6:05:50"}
+{"current_steps": 1040, "total_steps": 2726, "loss": 0.4661, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.417310344032309e-05, "epoch": 0.76, "percentage": 38.15, "elapsed_time": "3:44:11", "remaining_time": "6:03:26"}
+{"current_steps": 1050, "total_steps": 2726, "loss": 0.4945, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.390448212209191e-05, "epoch": 0.77, "percentage": 38.52, "elapsed_time": "3:46:22", "remaining_time": "6:01:20"}
+{"current_steps": 1060, "total_steps": 2726, "loss": 0.4399, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.3634678165663325e-05, "epoch": 0.78, "percentage": 38.88, "elapsed_time": "3:48:36", "remaining_time": "5:59:18"}
+{"current_steps": 1070, "total_steps": 2726, "loss": 0.4399, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.336372740472877e-05, "epoch": 0.78, "percentage": 39.25, "elapsed_time": "3:50:31", "remaining_time": "5:56:47"}
+{"current_steps": 1080, "total_steps": 2726, "loss": 0.4251, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.309166582529114e-05, "epoch": 0.79, "percentage": 39.62, "elapsed_time": "3:52:21", "remaining_time": "5:54:08"}
+{"current_steps": 1090, "total_steps": 2726, "loss": 0.4385, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.281852956088537e-05, "epoch": 0.8, "percentage": 39.99, "elapsed_time": "3:54:18", "remaining_time": "5:51:40"}
+{"current_steps": 1100, "total_steps": 2726, "loss": 0.4566, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.254435488777941e-05, "epoch": 0.81, "percentage": 40.35, "elapsed_time": "3:56:22", "remaining_time": "5:49:24"}
+{"current_steps": 1110, "total_steps": 2726, "loss": 0.4352, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.226917822015623e-05, "epoch": 0.81, "percentage": 40.72, "elapsed_time": "3:58:33", "remaining_time": "5:47:19"}
+{"current_steps": 1120, "total_steps": 2726, "loss": 0.4617, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.199303610527749e-05, "epoch": 0.82, "percentage": 41.09, "elapsed_time": "4:00:57", "remaining_time": "5:45:30"}
+{"current_steps": 1130, "total_steps": 2726, "loss": 0.476, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.1715965218629595e-05, "epoch": 0.83, "percentage": 41.45, "elapsed_time": "4:02:46", "remaining_time": "5:42:54"}
+{"current_steps": 1140, "total_steps": 2726, "loss": 0.4322, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.143800235905268e-05, "epoch": 0.84, "percentage": 41.82, "elapsed_time": "4:04:46", "remaining_time": "5:40:32"}
+{"current_steps": 1150, "total_steps": 2726, "loss": 0.4736, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.115918444385315e-05, "epoch": 0.84, "percentage": 42.19, "elapsed_time": "4:06:57", "remaining_time": "5:38:26"}
+{"current_steps": 1160, "total_steps": 2726, "loss": 0.4322, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.0879548503900665e-05, "epoch": 0.85, "percentage": 42.55, "elapsed_time": "4:09:21", "remaining_time": "5:36:37"}
+{"current_steps": 1170, "total_steps": 2726, "loss": 0.4152, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.0599131678709836e-05, "epoch": 0.86, "percentage": 42.92, "elapsed_time": "4:11:20", "remaining_time": "5:34:16"}
+{"current_steps": 1180, "total_steps": 2726, "loss": 0.4187, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.031797121150764e-05, "epoch": 0.87, "percentage": 43.29, "elapsed_time": "4:13:18", "remaining_time": "5:31:52"}
+{"current_steps": 1190, "total_steps": 2726, "loss": 0.4331, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.0036104444286954e-05, "epoch": 0.87, "percentage": 43.65, "elapsed_time": "4:15:22", "remaining_time": "5:29:37"}
+{"current_steps": 1200, "total_steps": 2726, "loss": 0.45, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.9753568812847065e-05, "epoch": 0.88, "percentage": 44.02, "elapsed_time": "4:17:56", "remaining_time": "5:28:00"}
+{"current_steps": 1200, "total_steps": 2726, "loss": null, "eval_loss": 0.43782830238342285, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 0.88, "percentage": 44.02, "elapsed_time": "4:17:56", "remaining_time": "5:28:00"}
+{"current_steps": 1210, "total_steps": 2726, "loss": 0.4457, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.9470401841821686e-05, "epoch": 0.89, "percentage": 44.39, "elapsed_time": "4:21:17", "remaining_time": "5:27:22"}
+{"current_steps": 1220, "total_steps": 2726, "loss": 0.4667, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.9186641139695108e-05, "epoch": 0.89, "percentage": 44.75, "elapsed_time": "4:23:21", "remaining_time": "5:25:05"}
+{"current_steps": 1230, "total_steps": 2726, "loss": 0.4516, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.8902324393807333e-05, "epoch": 0.9, "percentage": 45.12, "elapsed_time": "4:25:38", "remaining_time": "5:23:05"}
+{"current_steps": 1240, "total_steps": 2726, "loss": 0.4695, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.861748936534867e-05, "epoch": 0.91, "percentage": 45.49, "elapsed_time": "4:27:43", "remaining_time": "5:20:50"}
+{"current_steps": 1250, "total_steps": 2726, "loss": 0.45, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.8332173884344477e-05, "epoch": 0.92, "percentage": 45.85, "elapsed_time": "4:30:07", "remaining_time": "5:18:58"}
+{"current_steps": 1260, "total_steps": 2726, "loss": 0.4246, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.8046415844630857e-05, "epoch": 0.92, "percentage": 46.22, "elapsed_time": "4:32:04", "remaining_time": "5:16:33"}
+{"current_steps": 1270, "total_steps": 2726, "loss": 0.4449, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.7760253198821822e-05, "epoch": 0.93, "percentage": 46.59, "elapsed_time": "4:34:02", "remaining_time": "5:14:10"}
+{"current_steps": 1280, "total_steps": 2726, "loss": 0.4217, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.7473723953268687e-05, "epoch": 0.94, "percentage": 46.96, "elapsed_time": "4:36:21", "remaining_time": "5:12:11"}
+{"current_steps": 1290, "total_steps": 2726, "loss": 0.4495, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.7186866163012232e-05, "epoch": 0.95, "percentage": 47.32, "elapsed_time": "4:38:18", "remaining_time": "5:09:48"}
+{"current_steps": 1300, "total_steps": 2726, "loss": 0.4523, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.6899717926728535e-05, "epoch": 0.95, "percentage": 47.69, "elapsed_time": "4:40:31", "remaining_time": "5:07:43"}
+{"current_steps": 1310, "total_steps": 2726, "loss": 0.4523, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.6612317381668915e-05, "epoch": 0.96, "percentage": 48.06, "elapsed_time": "4:42:52", "remaining_time": "5:05:45"}
+{"current_steps": 1320, "total_steps": 2726, "loss": 0.45, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.632470269859478e-05, "epoch": 0.97, "percentage": 48.42, "elapsed_time": "4:44:43", "remaining_time": "5:03:16"}
+{"current_steps": 1330, "total_steps": 2726, "loss": 0.4371, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.603691207670803e-05, "epoch": 0.98, "percentage": 48.79, "elapsed_time": "4:46:50", "remaining_time": "5:01:04"}
+{"current_steps": 1340, "total_steps": 2726, "loss": 0.4221, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.5748983738577653e-05, "epoch": 0.98, "percentage": 49.16, "elapsed_time": "4:49:12", "remaining_time": "4:59:08"}
+{"current_steps": 1350, "total_steps": 2726, "loss": 0.4565, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.5460955925063268e-05, "epoch": 0.99, "percentage": 49.52, "elapsed_time": "4:51:37", "remaining_time": "4:57:14"}
+{"current_steps": 1360, "total_steps": 2726, "loss": 0.3892, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.5172866890236203e-05, "epoch": 1.0, "percentage": 49.89, "elapsed_time": "4:53:48", "remaining_time": "4:55:05"}
+{"current_steps": 1370, "total_steps": 2726, "loss": 0.3973, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.48847548962988e-05, "epoch": 1.0, "percentage": 50.26, "elapsed_time": "4:56:07", "remaining_time": "4:53:06"}
+{"current_steps": 1380, "total_steps": 2726, "loss": 0.4151, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.4596658208502713e-05, "epoch": 1.01, "percentage": 50.62, "elapsed_time": "4:58:09", "remaining_time": "4:50:48"}
+{"current_steps": 1390, "total_steps": 2726, "loss": 0.3839, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.4308615090066735e-05, "epoch": 1.02, "percentage": 50.99, "elapsed_time": "5:00:25", "remaining_time": "4:48:44"}
+{"current_steps": 1400, "total_steps": 2726, "loss": 0.4062, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.4020663797094864e-05, "epoch": 1.03, "percentage": 51.36, "elapsed_time": "5:02:30", "remaining_time": "4:46:31"}
+{"current_steps": 1400, "total_steps": 2726, "loss": null, "eval_loss": 0.43284282088279724, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 1.03, "percentage": 51.36, "elapsed_time": "5:02:30", "remaining_time": "4:46:31"}
+{"current_steps": 1410, "total_steps": 2726, "loss": 0.4003, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.373284257349544e-05, "epoch": 1.03, "percentage": 51.72, "elapsed_time": "5:05:58", "remaining_time": "4:45:34"}
+{"current_steps": 1420, "total_steps": 2726, "loss": 0.4447, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.3445189645901806e-05, "epoch": 1.04, "percentage": 52.09, "elapsed_time": "5:08:26", "remaining_time": "4:43:40"}
+{"current_steps": 1430, "total_steps": 2726, "loss": 0.4565, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.3157743218595247e-05, "epoch": 1.05, "percentage": 52.46, "elapsed_time": "5:10:20", "remaining_time": "4:41:15"}
+{"current_steps": 1440, "total_steps": 2726, "loss": 0.4265, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.287054146843097e-05, "epoch": 1.06, "percentage": 52.82, "elapsed_time": "5:12:13", "remaining_time": "4:38:50"}
+{"current_steps": 1450, "total_steps": 2726, "loss": 0.4056, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.2583622539767668e-05, "epoch": 1.06, "percentage": 53.19, "elapsed_time": "5:14:21", "remaining_time": "4:36:38"}
+{"current_steps": 1460, "total_steps": 2726, "loss": 0.4074, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.2297024539401463e-05, "epoch": 1.07, "percentage": 53.56, "elapsed_time": "5:16:26", "remaining_time": "4:34:23"}
+{"current_steps": 1470, "total_steps": 2726, "loss": 0.4281, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.2010785531504716e-05, "epoch": 1.08, "percentage": 53.93, "elapsed_time": "5:18:46", "remaining_time": "4:32:22"}
+{"current_steps": 1480, "total_steps": 2726, "loss": 0.4663, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.172494353257066e-05, "epoch": 1.09, "percentage": 54.29, "elapsed_time": "5:21:04", "remaining_time": "4:30:18"}
+{"current_steps": 1490, "total_steps": 2726, "loss": 0.4281, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.1439536506364274e-05, "epoch": 1.09, "percentage": 54.66, "elapsed_time": "5:23:36", "remaining_time": "4:28:26"}
+{"current_steps": 1500, "total_steps": 2726, "loss": 0.4111, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.1154602358880122e-05, "epoch": 1.1, "percentage": 55.03, "elapsed_time": "5:25:52", "remaining_time": "4:26:20"}
+{"current_steps": 1510, "total_steps": 2726, "loss": 0.4187, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.0870178933307948e-05, "epoch": 1.11, "percentage": 55.39, "elapsed_time": "5:27:49", "remaining_time": "4:23:59"}
+{"current_steps": 1520, "total_steps": 2726, "loss": 0.3964, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.0586304005006585e-05, "epoch": 1.11, "percentage": 55.76, "elapsed_time": "5:30:05", "remaining_time": "4:21:54"}
+{"current_steps": 1530, "total_steps": 2726, "loss": 0.4331, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.030301527648684e-05, "epoch": 1.12, "percentage": 56.13, "elapsed_time": "5:32:15", "remaining_time": "4:19:43"}
+{"current_steps": 1540, "total_steps": 2726, "loss": 0.4432, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.0020350372404102e-05, "epoch": 1.13, "percentage": 56.49, "elapsed_time": "5:34:15", "remaining_time": "4:17:25"}
+{"current_steps": 1550, "total_steps": 2726, "loss": 0.4056, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.9738346834561254e-05, "epoch": 1.14, "percentage": 56.86, "elapsed_time": "5:36:18", "remaining_time": "4:15:09"}
+{"current_steps": 1560, "total_steps": 2726, "loss": 0.4354, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.945704211692262e-05, "epoch": 1.14, "percentage": 57.23, "elapsed_time": "5:38:26", "remaining_time": "4:12:57"}
+{"current_steps": 1570, "total_steps": 2726, "loss": 0.4309, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.9176473580639538e-05, "epoch": 1.15, "percentage": 57.59, "elapsed_time": "5:40:42", "remaining_time": "4:10:52"}
+{"current_steps": 1580, "total_steps": 2726, "loss": 0.4058, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.8896678489088304e-05, "epoch": 1.16, "percentage": 57.96, "elapsed_time": "5:42:44", "remaining_time": "4:08:35"}
+{"current_steps": 1590, "total_steps": 2726, "loss": 0.4319, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.8617694002921064e-05, "epoch": 1.17, "percentage": 58.33, "elapsed_time": "5:44:54", "remaining_time": "4:06:25"}
+{"current_steps": 1600, "total_steps": 2726, "loss": 0.4267, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.8339557175130383e-05, "epoch": 1.17, "percentage": 58.69, "elapsed_time": "5:46:47", "remaining_time": "4:04:03"}
+{"current_steps": 1600, "total_steps": 2726, "loss": null, "eval_loss": 0.42894992232322693, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 1.17, "percentage": 58.69, "elapsed_time": "5:46:47", "remaining_time": "4:04:03"}
+{"current_steps": 1610, "total_steps": 2726, "loss": 0.3921, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.8062304946128073e-05, "epoch": 1.18, "percentage": 59.06, "elapsed_time": "5:50:11", "remaining_time": "4:02:44"}
+{"current_steps": 1620, "total_steps": 2726, "loss": 0.4206, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.7785974138839018e-05, "epoch": 1.19, "percentage": 59.43, "elapsed_time": "5:52:10", "remaining_time": "4:00:26"}
+{"current_steps": 1630, "total_steps": 2726, "loss": 0.4356, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.7510601453810594e-05, "epoch": 1.2, "percentage": 59.79, "elapsed_time": "5:54:17", "remaining_time": "3:58:13"}
+{"current_steps": 1640, "total_steps": 2726, "loss": 0.4167, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.723622346433828e-05, "epoch": 1.2, "percentage": 60.16, "elapsed_time": "5:56:29", "remaining_time": "3:56:03"}
+{"current_steps": 1650, "total_steps": 2726, "loss": 0.4233, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.6962876611608262e-05, "epoch": 1.21, "percentage": 60.53, "elapsed_time": "5:58:26", "remaining_time": "3:53:45"}
+{"current_steps": 1660, "total_steps": 2726, "loss": 0.4176, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.6690597199857523e-05, "epoch": 1.22, "percentage": 60.9, "elapsed_time": "6:00:33", "remaining_time": "3:51:32"}
+{"current_steps": 1670, "total_steps": 2726, "loss": 0.4241, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.6419421391552142e-05, "epoch": 1.22, "percentage": 61.26, "elapsed_time": "6:02:41", "remaining_time": "3:49:20"}
+{"current_steps": 1680, "total_steps": 2726, "loss": 0.4524, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.6149385202584423e-05, "epoch": 1.23, "percentage": 61.63, "elapsed_time": "6:04:45", "remaining_time": "3:47:06"}
+{"current_steps": 1690, "total_steps": 2726, "loss": 0.4647, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.5880524497489474e-05, "epoch": 1.24, "percentage": 62.0, "elapsed_time": "6:07:03", "remaining_time": "3:45:01"}
+{"current_steps": 1700, "total_steps": 2726, "loss": 0.4036, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.5612874984681923e-05, "epoch": 1.25, "percentage": 62.36, "elapsed_time": "6:09:03", "remaining_time": "3:42:44"}
+{"current_steps": 1710, "total_steps": 2726, "loss": 0.4148, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.534647221171334e-05, "epoch": 1.25, "percentage": 62.73, "elapsed_time": "6:11:22", "remaining_time": "3:40:38"}
+{"current_steps": 1720, "total_steps": 2726, "loss": 0.408, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.5081351560551021e-05, "epoch": 1.26, "percentage": 63.1, "elapsed_time": "6:13:19", "remaining_time": "3:38:21"}
+{"current_steps": 1730, "total_steps": 2726, "loss": 0.457, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.4817548242878759e-05, "epoch": 1.27, "percentage": 63.46, "elapsed_time": "6:15:23", "remaining_time": "3:36:07"}
+{"current_steps": 1740, "total_steps": 2726, "loss": 0.4193, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.45550972954203e-05, "epoch": 1.28, "percentage": 63.83, "elapsed_time": "6:17:21", "remaining_time": "3:33:50"}
+{"current_steps": 1750, "total_steps": 2726, "loss": 0.4149, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.4294033575285914e-05, "epoch": 1.28, "percentage": 64.2, "elapsed_time": "6:19:35", "remaining_time": "3:31:42"}
+{"current_steps": 1760, "total_steps": 2726, "loss": 0.4032, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.4034391755342972e-05, "epoch": 1.29, "percentage": 64.56, "elapsed_time": "6:21:52", "remaining_time": "3:29:35"}
+{"current_steps": 1770, "total_steps": 2726, "loss": 0.4078, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.3776206319610823e-05, "epoch": 1.3, "percentage": 64.93, "elapsed_time": "6:23:56", "remaining_time": "3:27:22"}
+{"current_steps": 1780, "total_steps": 2726, "loss": 0.4123, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.3519511558680892e-05, "epoch": 1.31, "percentage": 65.3, "elapsed_time": "6:26:23", "remaining_time": "3:25:21"}
+{"current_steps": 1790, "total_steps": 2726, "loss": 0.4089, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.3264341565162422e-05, "epoch": 1.31, "percentage": 65.66, "elapsed_time": "6:28:17", "remaining_time": "3:23:02"}
+{"current_steps": 1800, "total_steps": 2726, "loss": 0.383, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.3010730229154445e-05, "epoch": 1.32, "percentage": 66.03, "elapsed_time": "6:30:17", "remaining_time": "3:20:47"}
+{"current_steps": 1800, "total_steps": 2726, "loss": null, "eval_loss": 0.4250344932079315, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 1.32, "percentage": 66.03, "elapsed_time": "6:30:17", "remaining_time": "3:20:47"}
+{"current_steps": 1810, "total_steps": 2726, "loss": 0.3936, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.2758711233744783e-05, "epoch": 1.33, "percentage": 66.4, "elapsed_time": "6:33:42", "remaining_time": "3:19:14"}
+{"current_steps": 1820, "total_steps": 2726, "loss": 0.4127, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.2508318050536421e-05, "epoch": 1.33, "percentage": 66.76, "elapsed_time": "6:35:33", "remaining_time": "3:16:54"}
+{"current_steps": 1830, "total_steps": 2726, "loss": 0.4015, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.2259583935202062e-05, "epoch": 1.34, "percentage": 67.13, "elapsed_time": "6:37:28", "remaining_time": "3:14:36"}
+{"current_steps": 1840, "total_steps": 2726, "loss": 0.4203, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.2012541923067244e-05, "epoch": 1.35, "percentage": 67.5, "elapsed_time": "6:39:29", "remaining_time": "3:12:21"}
+{"current_steps": 1850, "total_steps": 2726, "loss": 0.4367, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.176722482472286e-05, "epoch": 1.36, "percentage": 67.87, "elapsed_time": "6:41:25", "remaining_time": "3:10:04"}
+{"current_steps": 1860, "total_steps": 2726, "loss": 0.4233, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.1523665221667398e-05, "epoch": 1.36, "percentage": 68.23, "elapsed_time": "6:43:12", "remaining_time": "3:07:44"}
+{"current_steps": 1870, "total_steps": 2726, "loss": 0.4405, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.1281895461979732e-05, "epoch": 1.37, "percentage": 68.6, "elapsed_time": "6:45:16", "remaining_time": "3:05:31"}
+{"current_steps": 1880, "total_steps": 2726, "loss": 0.4379, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.104194765602281e-05, "epoch": 1.38, "percentage": 68.97, "elapsed_time": "6:47:24", "remaining_time": "3:03:20"}
+{"current_steps": 1890, "total_steps": 2726, "loss": 0.4146, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.0803853672178946e-05, "epoch": 1.39, "percentage": 69.33, "elapsed_time": "6:49:44", "remaining_time": "3:01:14"}
+{"current_steps": 1900, "total_steps": 2726, "loss": 0.4438, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.0567645132617316e-05, "epoch": 1.39, "percentage": 69.7, "elapsed_time": "6:51:53", "remaining_time": "2:59:03"}
+{"current_steps": 1910, "total_steps": 2726, "loss": 0.3915, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.0333353409094015e-05, "epoch": 1.4, "percentage": 70.07, "elapsed_time": "6:54:02", "remaining_time": "2:56:53"}
+{"current_steps": 1920, "total_steps": 2726, "loss": 0.4063, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.0101009618785528e-05, "epoch": 1.41, "percentage": 70.43, "elapsed_time": "6:56:21", "remaining_time": "2:54:47"}
+{"current_steps": 1930, "total_steps": 2726, "loss": 0.3974, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.870644620155877e-06, "epoch": 1.42, "percentage": 70.8, "elapsed_time": "6:58:32", "remaining_time": "2:52:37"}
+{"current_steps": 1940, "total_steps": 2726, "loss": 0.4015, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.642289008858244e-06, "epoch": 1.42, "percentage": 71.17, "elapsed_time": "7:00:49", "remaining_time": "2:50:30"}
+{"current_steps": 1950, "total_steps": 2726, "loss": 0.4264, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.41597311367142e-06, "epoch": 1.43, "percentage": 71.53, "elapsed_time": "7:02:59", "remaining_time": "2:48:19"}
+{"current_steps": 1960, "total_steps": 2726, "loss": 0.4334, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.191726992471725e-06, "epoch": 1.44, "percentage": 71.9, "elapsed_time": "7:05:22", "remaining_time": "2:46:14"}
+{"current_steps": 1970, "total_steps": 2726, "loss": 0.4197, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 8.969580428240903e-06, "epoch": 1.44, "percentage": 72.27, "elapsed_time": "7:07:11", "remaining_time": "2:43:56"}
+{"current_steps": 1980, "total_steps": 2726, "loss": 0.422, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 8.74956292511056e-06, "epoch": 1.45, "percentage": 72.63, "elapsed_time": "7:09:23", "remaining_time": "2:41:46"}
+{"current_steps": 1990, "total_steps": 2726, "loss": 0.4077, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 8.531703704443575e-06, "epoch": 1.46, "percentage": 73.0, "elapsed_time": "7:11:33", "remaining_time": "2:39:36"}
+{"current_steps": 2000, "total_steps": 2726, "loss": 0.4334, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 8.316031700953086e-06, "epoch": 1.47, "percentage": 73.37, "elapsed_time": "7:13:58", "remaining_time": "2:37:31"}
+{"current_steps": 2000, "total_steps": 2726, "loss": null, "eval_loss": 0.4206145107746124, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 1.47, "percentage": 73.37, "elapsed_time": "7:13:58", "remaining_time": "2:37:31"}
+{"current_steps": 2010, "total_steps": 2726, "loss": 0.4171, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 8.102575558859612e-06, "epoch": 1.47, "percentage": 73.73, "elapsed_time": "7:17:28", "remaining_time": "2:35:50"}
+{"current_steps": 2020, "total_steps": 2726, "loss": 0.3957, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 7.891363628086671e-06, "epoch": 1.48, "percentage": 74.1, "elapsed_time": "7:19:27", "remaining_time": "2:33:35"}
+{"current_steps": 2030, "total_steps": 2726, "loss": 0.4174, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 7.682423960495517e-06, "epoch": 1.49, "percentage": 74.47, "elapsed_time": "7:21:25", "remaining_time": "2:31:20"}
+{"current_steps": 2040, "total_steps": 2726, "loss": 0.4187, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 7.475784306159478e-06, "epoch": 1.5, "percentage": 74.83, "elapsed_time": "7:23:33", "remaining_time": "2:29:09"}
+{"current_steps": 2050, "total_steps": 2726, "loss": 0.3775, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 7.271472109678379e-06, "epoch": 1.5, "percentage": 75.2, "elapsed_time": "7:25:41", "remaining_time": "2:26:58"}
+{"current_steps": 2060, "total_steps": 2726, "loss": 0.4019, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 7.0695145065334585e-06, "epoch": 1.51, "percentage": 75.57, "elapsed_time": "7:27:36", "remaining_time": "2:24:42"}
+{"current_steps": 2070, "total_steps": 2726, "loss": 0.4143, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.869938319483471e-06, "epoch": 1.52, "percentage": 75.94, "elapsed_time": "7:29:45", "remaining_time": "2:22:31"}
+{"current_steps": 2080, "total_steps": 2726, "loss": 0.3853, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.67277005500222e-06, "epoch": 1.53, "percentage": 76.3, "elapsed_time": "7:31:46", "remaining_time": "2:20:18"}
+{"current_steps": 2090, "total_steps": 2726, "loss": 0.4403, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.478035899758139e-06, "epoch": 1.53, "percentage": 76.67, "elapsed_time": "7:33:54", "remaining_time": "2:18:07"}
+{"current_steps": 2100, "total_steps": 2726, "loss": 0.3888, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.285761717136335e-06, "epoch": 1.54, "percentage": 77.04, "elapsed_time": "7:36:12", "remaining_time": "2:15:59"}
+{"current_steps": 2110, "total_steps": 2726, "loss": 0.4123, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.095973043803577e-06, "epoch": 1.55, "percentage": 77.4, "elapsed_time": "7:38:32", "remaining_time": "2:13:52"}
+{"current_steps": 2120, "total_steps": 2726, "loss": 0.4035, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.908695086316701e-06, "epoch": 1.55, "percentage": 77.77, "elapsed_time": "7:40:48", "remaining_time": "2:11:43"}
+{"current_steps": 2130, "total_steps": 2726, "loss": 0.4219, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.723952717774763e-06, "epoch": 1.56, "percentage": 78.14, "elapsed_time": "7:42:48", "remaining_time": "2:09:30"}
+{"current_steps": 2140, "total_steps": 2726, "loss": 0.4209, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.54177047451562e-06, "epoch": 1.57, "percentage": 78.5, "elapsed_time": "7:44:53", "remaining_time": "2:07:18"}
+{"current_steps": 2150, "total_steps": 2726, "loss": 0.4286, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.362172552857128e-06, "epoch": 1.58, "percentage": 78.87, "elapsed_time": "7:47:02", "remaining_time": "2:05:07"}
+{"current_steps": 2160, "total_steps": 2726, "loss": 0.4134, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.18518280588354e-06, "epoch": 1.58, "percentage": 79.24, "elapsed_time": "7:49:04", "remaining_time": "2:02:54"}
+{"current_steps": 2170, "total_steps": 2726, "loss": 0.4005, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.010824740277501e-06, "epoch": 1.59, "percentage": 79.6, "elapsed_time": "7:51:12", "remaining_time": "2:00:44"}
+{"current_steps": 2180, "total_steps": 2726, "loss": 0.3973, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.83912151319802e-06, "epoch": 1.6, "percentage": 79.97, "elapsed_time": "7:53:12", "remaining_time": "1:58:31"}
+{"current_steps": 2190, "total_steps": 2726, "loss": 0.3959, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.6700959292048875e-06, "epoch": 1.61, "percentage": 80.34, "elapsed_time": "7:55:29", "remaining_time": "1:56:22"}
+{"current_steps": 2200, "total_steps": 2726, "loss": 0.4017, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.503770437229918e-06, "epoch": 1.61, "percentage": 80.7, "elapsed_time": "7:57:38", "remaining_time": "1:54:12"}
+{"current_steps": 2200, "total_steps": 2726, "loss": null, "eval_loss": 0.41970890760421753, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 1.61, "percentage": 80.7, "elapsed_time": "7:57:38", "remaining_time": "1:54:12"}
+{"current_steps": 2210, "total_steps": 2726, "loss": 0.4013, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.340167127595407e-06, "epoch": 1.62, "percentage": 81.07, "elapsed_time": "8:00:51", "remaining_time": "1:52:16"}
+{"current_steps": 2220, "total_steps": 2726, "loss": 0.4069, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.179307729080256e-06, "epoch": 1.63, "percentage": 81.44, "elapsed_time": "8:03:00", "remaining_time": "1:50:05"}
+{"current_steps": 2230, "total_steps": 2726, "loss": 0.3857, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.021213606034064e-06, "epoch": 1.64, "percentage": 81.8, "elapsed_time": "8:05:05", "remaining_time": "1:47:53"}
+{"current_steps": 2240, "total_steps": 2726, "loss": 0.4262, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.8659057555396645e-06, "epoch": 1.64, "percentage": 82.17, "elapsed_time": "8:07:03", "remaining_time": "1:45:40"}
+{"current_steps": 2250, "total_steps": 2726, "loss": 0.3953, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.713404804624418e-06, "epoch": 1.65, "percentage": 82.54, "elapsed_time": "8:09:07", "remaining_time": "1:43:28"}
+{"current_steps": 2260, "total_steps": 2726, "loss": 0.4231, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.5637310075206544e-06, "epoch": 1.66, "percentage": 82.91, "elapsed_time": "8:11:23", "remaining_time": "1:41:19"}
+{"current_steps": 2270, "total_steps": 2726, "loss": 0.407, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.416904242975635e-06, "epoch": 1.66, "percentage": 83.27, "elapsed_time": "8:13:23", "remaining_time": "1:39:06"}
+{"current_steps": 2280, "total_steps": 2726, "loss": 0.4254, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.2729440116113843e-06, "epoch": 1.67, "percentage": 83.64, "elapsed_time": "8:15:40", "remaining_time": "1:36:57"}
+{"current_steps": 2290, "total_steps": 2726, "loss": 0.4054, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.131869433334725e-06, "epoch": 1.68, "percentage": 84.01, "elapsed_time": "8:17:43", "remaining_time": "1:34:45"}
+{"current_steps": 2300, "total_steps": 2726, "loss": 0.3801, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.9936992447979068e-06, "epoch": 1.69, "percentage": 84.37, "elapsed_time": "8:20:05", "remaining_time": "1:32:37"}
+{"current_steps": 2310, "total_steps": 2726, "loss": 0.4335, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.8584517969101054e-06, "epoch": 1.69, "percentage": 84.74, "elapsed_time": "8:21:46", "remaining_time": "1:30:21"}
+{"current_steps": 2320, "total_steps": 2726, "loss": 0.412, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.7261450524001807e-06, "epoch": 1.7, "percentage": 85.11, "elapsed_time": "8:23:37", "remaining_time": "1:28:08"}
+{"current_steps": 2330, "total_steps": 2726, "loss": 0.4456, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.596796583430969e-06, "epoch": 1.71, "percentage": 85.47, "elapsed_time": "8:25:39", "remaining_time": "1:25:56"}
+{"current_steps": 2340, "total_steps": 2726, "loss": 0.3972, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.470423569265462e-06, "epoch": 1.72, "percentage": 85.84, "elapsed_time": "8:27:30", "remaining_time": "1:23:43"}
+{"current_steps": 2350, "total_steps": 2726, "loss": 0.3935, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.34704279398516e-06, "epoch": 1.72, "percentage": 86.21, "elapsed_time": "8:29:12", "remaining_time": "1:21:28"}
+{"current_steps": 2360, "total_steps": 2726, "loss": 0.3974, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.2266706442609226e-06, "epoch": 1.73, "percentage": 86.57, "elapsed_time": "8:31:13", "remaining_time": "1:19:17"}
+{"current_steps": 2370, "total_steps": 2726, "loss": 0.4157, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.109323107176578e-06, "epoch": 1.74, "percentage": 86.94, "elapsed_time": "8:33:31", "remaining_time": "1:17:08"}
+{"current_steps": 2380, "total_steps": 2726, "loss": 0.4379, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.9950157681056318e-06, "epoch": 1.75, "percentage": 87.31, "elapsed_time": "8:35:40", "remaining_time": "1:14:58"}
+{"current_steps": 2390, "total_steps": 2726, "loss": 0.4035, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.8837638086413063e-06, "epoch": 1.75, "percentage": 87.67, "elapsed_time": "8:37:35", "remaining_time": "1:12:46"}
+{"current_steps": 2400, "total_steps": 2726, "loss": 0.4455, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.7755820045802145e-06, "epoch": 1.76, "percentage": 88.04, "elapsed_time": "8:39:45", "remaining_time": "1:10:35"}
+{"current_steps": 2400, "total_steps": 2726, "loss": null, "eval_loss": 0.41885045170783997, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 1.76, "percentage": 88.04, "elapsed_time": "8:39:45", "remaining_time": "1:10:35"}
+{"current_steps": 2410, "total_steps": 2726, "loss": 0.4532, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.6704847239599364e-06, "epoch": 1.77, "percentage": 88.41, "elapsed_time": "8:43:09", "remaining_time": "1:08:35"}
+{"current_steps": 2420, "total_steps": 2726, "loss": 0.4213, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.5684859251507394e-06, "epoch": 1.77, "percentage": 88.77, "elapsed_time": "8:45:22", "remaining_time": "1:06:25"}
+{"current_steps": 2430, "total_steps": 2726, "loss": 0.3873, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.4695991550017164e-06, "epoch": 1.78, "percentage": 89.14, "elapsed_time": "8:47:29", "remaining_time": "1:04:15"}
+{"current_steps": 2440, "total_steps": 2726, "loss": 0.4163, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.373837547041576e-06, "epoch": 1.79, "percentage": 89.51, "elapsed_time": "8:49:41", "remaining_time": "1:02:05"}
+{"current_steps": 2450, "total_steps": 2726, "loss": 0.3946, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.2812138197343392e-06, "epoch": 1.8, "percentage": 89.88, "elapsed_time": "8:51:48", "remaining_time": "0:59:54"}
+{"current_steps": 2460, "total_steps": 2726, "loss": 0.41, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.1917402747901152e-06, "epoch": 1.8, "percentage": 90.24, "elapsed_time": "8:53:44", "remaining_time": "0:57:42"}
+{"current_steps": 2470, "total_steps": 2726, "loss": 0.4126, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.105428795531327e-06, "epoch": 1.81, "percentage": 90.61, "elapsed_time": "8:55:59", "remaining_time": "0:55:33"}
+{"current_steps": 2480, "total_steps": 2726, "loss": 0.4505, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.0222908453143804e-06, "epoch": 1.82, "percentage": 90.98, "elapsed_time": "8:58:09", "remaining_time": "0:53:22"}
+{"current_steps": 2490, "total_steps": 2726, "loss": 0.4303, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.423374660072065e-07, "epoch": 1.83, "percentage": 91.34, "elapsed_time": "9:00:00", "remaining_time": "0:51:10"}
+{"current_steps": 2500, "total_steps": 2726, "loss": 0.4179, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 8.655792765227405e-07, "epoch": 1.83, "percentage": 91.71, "elapsed_time": "9:01:58", "remaining_time": "0:48:59"}
+{"current_steps": 2510, "total_steps": 2726, "loss": 0.4376, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 7.920264714085828e-07, "epoch": 1.84, "percentage": 92.08, "elapsed_time": "9:04:05", "remaining_time": "0:46:49"}
+{"current_steps": 2520, "total_steps": 2726, "loss": 0.4069, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 7.216888194930272e-07, "epoch": 1.85, "percentage": 92.44, "elapsed_time": "9:06:08", "remaining_time": "0:44:38"}
+{"current_steps": 2530, "total_steps": 2726, "loss": 0.3887, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.545756625876031e-07, "epoch": 1.86, "percentage": 92.81, "elapsed_time": "9:08:12", "remaining_time": "0:42:28"}
+{"current_steps": 2540, "total_steps": 2726, "loss": 0.3972, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.906959142463947e-07, "epoch": 1.86, "percentage": 93.18, "elapsed_time": "9:10:08", "remaining_time": "0:40:17"}
+{"current_steps": 2550, "total_steps": 2726, "loss": 0.4078, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.300580585821696e-07, "epoch": 1.87, "percentage": 93.54, "elapsed_time": "9:12:17", "remaining_time": "0:38:07"}
+{"current_steps": 2560, "total_steps": 2726, "loss": 0.4467, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.7267014913956463e-07, "epoch": 1.88, "percentage": 93.91, "elapsed_time": "9:14:11", "remaining_time": "0:35:56"}
+{"current_steps": 2570, "total_steps": 2726, "loss": 0.4567, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.1853980782549097e-07, "epoch": 1.88, "percentage": 94.28, "elapsed_time": "9:16:09", "remaining_time": "0:33:45"}
+{"current_steps": 2580, "total_steps": 2726, "loss": 0.4494, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.6767422389682173e-07, "epoch": 1.89, "percentage": 94.64, "elapsed_time": "9:18:36", "remaining_time": "0:31:36"}
+{"current_steps": 2590, "total_steps": 2726, "loss": 0.4344, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.2008015300555306e-07, "epoch": 1.9, "percentage": 95.01, "elapsed_time": "9:20:48", "remaining_time": "0:29:26"}
+{"current_steps": 2600, "total_steps": 2726, "loss": 0.4101, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.757639163015774e-07, "epoch": 1.91, "percentage": 95.38, "elapsed_time": "9:22:52", "remaining_time": "0:27:16"}
+{"current_steps": 2600, "total_steps": 2726, "loss": null, "eval_loss": 0.41886425018310547, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 1.91, "percentage": 95.38, "elapsed_time": "9:22:52", "remaining_time": "0:27:16"}
+{"current_steps": 2610, "total_steps": 2726, "loss": 0.4058, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.347313995931466e-07, "epoch": 1.91, "percentage": 95.74, "elapsed_time": "9:26:02", "remaining_time": "0:25:09"}
+{"current_steps": 2620, "total_steps": 2726, "loss": 0.414, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.9698805256513908e-07, "epoch": 1.92, "percentage": 96.11, "elapsed_time": "9:28:14", "remaining_time": "0:22:59"}
+{"current_steps": 2630, "total_steps": 2726, "loss": 0.4254, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.6253888805527474e-07, "epoch": 1.93, "percentage": 96.48, "elapsed_time": "9:30:26", "remaining_time": "0:20:49"}
+{"current_steps": 2640, "total_steps": 2726, "loss": 0.4017, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.3138848138835313e-07, "epoch": 1.94, "percentage": 96.85, "elapsed_time": "9:32:34", "remaining_time": "0:18:39"}
+{"current_steps": 2650, "total_steps": 2726, "loss": 0.4258, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.0354096976856186e-07, "epoch": 1.94, "percentage": 97.21, "elapsed_time": "9:34:48", "remaining_time": "0:16:29"}
+{"current_steps": 2660, "total_steps": 2726, "loss": 0.4098, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 7.900005173002712e-08, "epoch": 1.95, "percentage": 97.58, "elapsed_time": "9:37:06", "remaining_time": "0:14:19"}
+{"current_steps": 2670, "total_steps": 2726, "loss": 0.42, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.776898664557051e-08, "epoch": 1.96, "percentage": 97.95, "elapsed_time": "9:39:14", "remaining_time": "0:12:08"}
+{"current_steps": 2680, "total_steps": 2726, "loss": 0.3886, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.985059429383875e-08, "epoch": 1.97, "percentage": 98.31, "elapsed_time": "9:41:31", "remaining_time": "0:09:58"}
+{"current_steps": 2690, "total_steps": 2726, "loss": 0.4276, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.5247254484794813e-08, "epoch": 1.97, "percentage": 98.68, "elapsed_time": "9:43:48", "remaining_time": "0:07:48"}
+{"current_steps": 2700, "total_steps": 2726, "loss": 0.4006, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.3960906743634706e-08, "epoch": 1.98, "percentage": 99.05, "elapsed_time": "9:45:40", "remaining_time": "0:05:38"}
+{"current_steps": 2710, "total_steps": 2726, "loss": 0.4243, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.993050053204607e-09, "epoch": 1.99, "percentage": 99.41, "elapsed_time": "9:47:47", "remaining_time": "0:03:28"}
+{"current_steps": 2720, "total_steps": 2726, "loss": 0.4439, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.3447426549129117e-09, "epoch": 1.99, "percentage": 99.78, "elapsed_time": "9:50:02", "remaining_time": "0:01:18"}
+{"current_steps": 2726, "total_steps": 2726, "loss": null, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 2.0, "percentage": 100.0, "elapsed_time": "9:51:16", "remaining_time": "0:00:00"}
+{"current_steps": 111, "total_steps": 111, "loss": null, "eval_loss": 0.41885045170783997, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 2.0, "percentage": 100.0, "elapsed_time": "9:52:25", "remaining_time": "0:00:00"}

{Baichuan-13B-Chat-full → Baichuan-13B-Chat-lora-Task}/trainer_state.json

@@ -1,1207 +1,1761 @@
 {
-  "best_metric": null,
-  "best_model_checkpoint": null,
-  "epoch": 2.0,
-  "global_step": 1974,
+  "best_metric": 0.41885045170783997,
+  "best_model_checkpoint": "output/Baichuan-13B-Chat_lora_wqs_nlp/checkpoint-2400",
+  "epoch": 1.9992665933259994,
+  "global_step": 2726,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.01,
-      "learning_rate": 4.9996834033646177e-05,
-      "loss": 0.9175,
+      "learning_rate": 4.999865525734509e-05,
+      "loss": 1.5452,
       "step": 10
     },
     {
-      "epoch": 0.02,
-      "learning_rate": 4.998733693645213e-05,
-      "loss": 0.7595,
+      "epoch": 0.01,
+      "learning_rate": 4.9994621174046976e-05,
+      "loss": 1.3291,
       "step": 20
     },
     {
-      "epoch": 0.03,
-      "learning_rate": 4.997151111381707e-05,
-      "loss": 0.7375,
+      "epoch": 0.02,
+      "learning_rate": 4.9986985231938546e-05,
+      "loss": 1.1048,
       "step": 30
     },
     {
-      "epoch": 0.04,
-      "learning_rate": 4.9949360574062774e-05,
-      "loss": 0.7227,
+      "epoch": 0.03,
+      "learning_rate": 4.99760306731191e-05,
+      "loss": 0.9155,
       "step": 40
     },
     {
-      "epoch": 0.05,
-      "learning_rate": 4.9920890927418316e-05,
-      "loss": 0.7147,
+      "epoch": 0.04,
+      "learning_rate": 4.9961758952505326e-05,
+      "loss": 0.8856,
       "step": 50
     },
     {
-      "epoch": 0.06,
-      "learning_rate": 4.988610938459917e-05,
-      "loss": 0.7102,
+      "epoch": 0.04,
+      "learning_rate": 4.9944171965578836e-05,
+      "loss": 0.8052,
       "step": 60
     },
     {
-      "epoch": 0.07,
-      "learning_rate": 4.9845024754980876e-05,
-      "loss": 0.7056,
+      "epoch": 0.05,
+      "learning_rate": 4.992327204813435e-05,
+      "loss": 0.7332,
       "step": 70
     },
     {
-      "epoch": 0.08,
-      "learning_rate": 4.979764744436784e-05,
-      "loss": 0.7128,
+      "epoch": 0.06,
+      "learning_rate": 4.989906197596955e-05,
+      "loss": 0.6982,
       "step": 80
     },
     {
-      "epoch": 0.09,
-      "learning_rate": 4.9743989452357756e-05,
-      "loss": 0.6982,
+      "epoch": 0.07,
+      "learning_rate": 4.987154496451635e-05,
+      "loss": 0.6811,
       "step": 90
     },
     {
-      "epoch": 0.1,
-      "learning_rate": 4.968406436930243e-05,
-      "loss": 0.7258,
+      "epoch": 0.07,
+      "learning_rate": 4.984072466841389e-05,
+      "loss": 0.6323,
       "step": 100
     },
     {
-      "epoch": 0.11,
-      "learning_rate": 4.961788737286559e-05,
-      "loss": 0.7095,
+      "epoch": 0.08,
+      "learning_rate": 4.981016546765289e-05,
+      "loss": 0.6234,
       "step": 110
     },
     {
-      "epoch": 0.12,
-      "learning_rate": 4.954547522417877e-05,
-      "loss": 0.7048,
+      "epoch": 0.09,
+      "learning_rate": 4.977308057009269e-05,
+      "loss": 0.6125,
       "step": 120
     },
     {
-      "epoch": 0.13,
-      "learning_rate": 4.946684626359607e-05,
-      "loss": 0.6805,
+      "epoch": 0.1,
+      "learning_rate": 4.97327054653146e-05,
+      "loss": 0.5977,
       "step": 130
     },
     {
-      "epoch": 0.14,
-      "learning_rate": 4.938202040604898e-05,
-      "loss": 0.6798,
+      "epoch": 0.1,
+      "learning_rate": 4.968904551569013e-05,
+      "loss": 0.6179,
       "step": 140
     },
     {
-      "epoch": 0.15,
-      "learning_rate": 4.929101913600238e-05,
-      "loss": 0.7134,
+      "epoch": 0.11,
+      "learning_rate": 4.9642106519863544e-05,
+      "loss": 0.5879,
       "step": 150
     },
     {
-      "epoch": 0.16,
-      "learning_rate": 4.919386550201299e-05,
-      "loss": 0.6895,
+      "epoch": 0.12,
+      "learning_rate": 4.959189471198171e-05,
+      "loss": 0.5472,
       "step": 160
     },
     {
-      "epoch": 0.17,
-      "learning_rate": 4.909058411089174e-05,
-      "loss": 0.705,
+      "epoch": 0.12,
+      "learning_rate": 4.953841676086613e-05,
+      "loss": 0.602,
       "step": 170
     },
     {
-      "epoch": 0.18,
-      "learning_rate": 4.8981201121471356e-05,
-      "loss": 0.6712,
+      "epoch": 0.13,
+      "learning_rate": 4.9481679769127275e-05,
+      "loss": 0.5773,
       "step": 180
     },
     {
-      "epoch": 0.19,
-      "learning_rate": 4.886574423798097e-05,
-      "loss": 0.6744,
+      "epoch": 0.14,
+      "learning_rate": 4.9421691272221167e-05,
+      "loss": 0.5615,
       "step": 190
     },
     {
-      "epoch": 0.2,
-      "learning_rate": 4.874424270302927e-05,
-      "loss": 0.6675,
+      "epoch": 0.15,
+      "learning_rate": 4.935845923744865e-05,
+      "loss": 0.5704,
       "step": 200
     },
     {
-      "epoch": 0.21,
-      "learning_rate": 4.861672729019797e-05,
-      "loss": 0.6726,
+      "epoch": 0.15,
+      "eval_loss": 0.5564362406730652,
+      "eval_runtime": 72.9469,
+      "eval_samples_per_second": 15.107,
+      "eval_steps_per_second": 1.522,
+      "step": 200
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 4.9291992062897183e-05,
+      "loss": 0.5803,
       "step": 210
     },
     {
-      "epoch": 0.22,
-      "learning_rate": 4.848323029624761e-05,
-      "loss": 0.6821,
+      "epoch": 0.16,
+      "learning_rate": 4.922229857632545e-05,
+      "loss": 0.5655,
       "step": 220
     },
     {
-      "epoch": 0.23,
-      "learning_rate": 4.834378553293748e-05,
-      "loss": 0.7133,
+      "epoch": 0.17,
+      "learning_rate": 4.9149388033990966e-05,
+      "loss": 0.5769,
       "step": 230
     },
     {
-      "epoch": 0.24,
-      "learning_rate": 4.81984283184619e-05,
-      "loss": 0.6681,
+      "epoch": 0.18,
+      "learning_rate": 4.9073270119420635e-05,
+      "loss": 0.5862,
       "step": 240
     },
     {
-      "epoch": 0.25,
-      "learning_rate": 4.804719546850487e-05,
-      "loss": 0.682,
+      "epoch": 0.18,
+      "learning_rate": 4.899395494212471e-05,
+      "loss": 0.5506,
       "step": 250
     },
     {
-      "epoch": 0.26,
-      "learning_rate": 4.789012528691558e-05,
-      "loss": 0.6755,
+      "epoch": 0.19,
+      "learning_rate": 4.891145303625408e-05,
+      "loss": 0.5183,
       "step": 260
     },
     {
-      "epoch": 0.27,
-      "learning_rate": 4.772725755600682e-05,
-      "loss": 0.68,
+      "epoch": 0.2,
+      "learning_rate": 4.882577535920121e-05,
+      "loss": 0.575,
       "step": 270
     },
     {
-      "epoch": 0.28,
-      "learning_rate": 4.755863352647909e-05,
-      "loss": 0.6663,
+      "epoch": 0.21,
+      "learning_rate": 4.8736933290144815e-05,
+      "loss": 0.5359,
       "step": 280
     },
     {
-      "epoch": 0.29,
-      "learning_rate": 4.738429590697271e-05,
-      "loss": 0.6672,
+      "epoch": 0.21,
+      "learning_rate": 4.8644938628538606e-05,
+      "loss": 0.5302,
       "step": 290
     },
     {
-      "epoch": 0.3,
-      "learning_rate": 4.720428885325069e-05,
-      "loss": 0.6466,
+      "epoch": 0.22,
+      "learning_rate": 4.8549803592544076e-05,
+      "loss": 0.5399,
       "step": 300
     },
     {
-      "epoch": 0.31,
-      "learning_rate": 4.701865795701505e-05,
-      "loss": 0.6668,
+      "epoch": 0.23,
+      "learning_rate": 4.845154081740783e-05,
+      "loss": 0.5379,
       "step": 310
     },
     {
-      "epoch": 0.32,
-      "learning_rate": 4.684682059461469e-05,
-      "loss": 0.6591,
+      "epoch": 0.23,
+      "learning_rate": 4.835016335378343e-05,
+      "loss": 0.5299,
       "step": 320
     },
     {
-      "epoch": 0.33,
-      "learning_rate": 4.665063509461097e-05,
-      "loss": 0.661,
+      "epoch": 0.24,
+      "learning_rate": 4.8245684665998073e-05,
+      "loss": 0.535,
       "step": 330
     },
     {
-      "epoch": 0.34,
-      "learning_rate": 4.644896598002736e-05,
-      "loss": 0.6749,
+      "epoch": 0.25,
+      "learning_rate": 4.813811863026436e-05,
+      "loss": 0.5145,
       "step": 340
     },
     {
-      "epoch": 0.35,
-      "learning_rate": 4.624186432907437e-05,
-      "loss": 0.6654,
+      "epoch": 0.26,
+      "learning_rate": 4.802747953283732e-05,
+      "loss": 0.5126,
       "step": 350
     },
     {
-      "epoch": 0.36,
-      "learning_rate": 4.602938259590072e-05,
-      "loss": 0.6716,
+      "epoch": 0.26,
+      "learning_rate": 4.791378206811704e-05,
+      "loss": 0.5012,
       "step": 360
     },
     {
-      "epoch": 0.37,
-      "learning_rate": 4.581157459730783e-05,
-      "loss": 0.6796,
+      "epoch": 0.27,
+      "learning_rate": 4.7797041336696995e-05,
+      "loss": 0.5373,
       "step": 370
     },
     {
-      "epoch": 0.39,
-      "learning_rate": 4.558849549911931e-05,
-      "loss": 0.6794,
+      "epoch": 0.28,
+      "learning_rate": 4.767727284335852e-05,
+      "loss": 0.4778,
       "step": 380
     },
     {
-      "epoch": 0.4,
-      "learning_rate": 4.536020180220871e-05,
-      "loss": 0.6651,
+      "epoch": 0.29,
+      "learning_rate": 4.755449249501155e-05,
+      "loss": 0.4894,
       "step": 390
     },
     {
-      "epoch": 0.41,
-      "learning_rate": 4.515032675559024e-05,
-      "loss": 0.6823,
+      "epoch": 0.29,
+      "learning_rate": 4.7428716598581934e-05,
+      "loss": 0.4986,
       "step": 400
     },
     {
-      "epoch": 0.42,
-      "learning_rate": 4.4912285699446786e-05,
-      "loss": 0.6677,
+      "epoch": 0.29,
+      "eval_loss": 0.4967592656612396,
+      "eval_runtime": 70.0948,
+      "eval_samples_per_second": 15.722,
+      "eval_steps_per_second": 1.584,
+      "step": 400
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 4.729996185884571e-05,
+      "loss": 0.4928,
       "step": 410
     },
     {
-      "epoch": 0.43,
-      "learning_rate": 4.4669201313179155e-05,
-      "loss": 0.6704,
+      "epoch": 0.31,
+      "learning_rate": 4.716824537621042e-05,
+      "loss": 0.4807,
       "step": 420
     },
     {
-      "epoch": 0.44,
-      "learning_rate": 4.442113516454638e-05,
-      "loss": 0.6481,
+      "epoch": 0.32,
+      "learning_rate": 4.703358464444397e-05,
+      "loss": 0.4734,
       "step": 430
     },
     {
-      "epoch": 0.45,
-      "learning_rate": 4.416815008307488e-05,
-      "loss": 0.665,
+      "epoch": 0.32,
+      "learning_rate": 4.689599754835122e-05,
+      "loss": 0.4603,
       "step": 440
     },
     {
-      "epoch": 0.46,
-      "learning_rate": 4.391031014414514e-05,
-      "loss": 0.6658,
+      "epoch": 0.33,
+      "learning_rate": 4.6755502361398616e-05,
+      "loss": 0.4661,
       "step": 450
     },
     {
-      "epoch": 0.47,
-      "learning_rate": 4.364768065276284e-05,
-      "loss": 0.6699,
+      "epoch": 0.34,
+      "learning_rate": 4.6612117743287234e-05,
+      "loss": 0.5072,
       "step": 460
     },
     {
-      "epoch": 0.48,
-      "learning_rate": 4.338032812701867e-05,
-      "loss": 0.6664,
+      "epoch": 0.34,
+      "learning_rate": 4.646586273747452e-05,
+      "loss": 0.5056,
       "step": 470
     },
     {
-      "epoch": 0.49,
-      "learning_rate": 4.310832028124069e-05,
-      "loss": 0.6817,
+      "epoch": 0.35,
+      "learning_rate": 4.631675676864503e-05,
+      "loss": 0.4391,
       "step": 480
     },
     {
-      "epoch": 0.5,
-      "learning_rate": 4.283172600884393e-05,
-      "loss": 0.6791,
+      "epoch": 0.36,
+      "learning_rate": 4.6164819640130595e-05,
+      "loss": 0.483,
       "step": 490
     },
     {
-      "epoch": 0.51,
-      "learning_rate": 4.2550615364881194e-05,
-      "loss": 0.6679,
+      "epoch": 0.37,
+      "learning_rate": 4.601007153128014e-05,
+      "loss": 0.4774,
       "step": 500
     },
     {
-      "epoch": 0.52,
-      "learning_rate": 4.226505954829973e-05,
-      "loss": 0.6476,
+      "epoch": 0.37,
+      "learning_rate": 4.5852532994779606e-05,
+      "loss": 0.4848,
       "step": 510
     },
     {
-      "epoch": 0.53,
-      "learning_rate": 4.197513088390813e-05,
-      "loss": 0.6721,
+      "epoch": 0.38,
+      "learning_rate": 4.5692224953922266e-05,
+      "loss": 0.5099,
       "step": 520
     },
     {
-      "epoch": 0.54,
-      "learning_rate": 4.1680902804058095e-05,
-      "loss": 0.6563,
+      "epoch": 0.39,
+      "learning_rate": 4.5529168699829805e-05,
+      "loss": 0.4851,
       "step": 530
     },
     {
-      "epoch": 0.55,
-      "learning_rate": 4.138244983004574e-05,
-      "loss": 0.6391,
+      "epoch": 0.4,
+      "learning_rate": 4.536338588862459e-05,
+      "loss": 0.498,
       "step": 540
     },
     {
-      "epoch": 0.56,
-      "learning_rate": 4.107984755323697e-05,
-      "loss": 0.6672,
+      "epoch": 0.4,
+      "learning_rate": 4.519489853855341e-05,
+      "loss": 0.5398,
       "step": 550
     },
     {
-      "epoch": 0.57,
-      "learning_rate": 4.077317261592194e-05,
-      "loss": 0.6497,
+      "epoch": 0.41,
+      "learning_rate": 4.50237290270632e-05,
+      "loss": 0.4566,
       "step": 560
     },
     {
-      "epoch": 0.58,
-      "learning_rate": 4.04625026919033e-05,
-      "loss": 0.668,
+      "epoch": 0.42,
+      "learning_rate": 4.484990008782893e-05,
+      "loss": 0.4942,
       "step": 570
     },
     {
-      "epoch": 0.59,
-      "learning_rate": 4.0147916466823174e-05,
-      "loss": 0.6682,
+      "epoch": 0.43,
+      "learning_rate": 4.467343480773433e-05,
+      "loss": 0.4625,
       "step": 580
     },
     {
-      "epoch": 0.6,
-      "learning_rate": 3.982949361823388e-05,
-      "loss": 0.6352,
+      "epoch": 0.43,
+      "learning_rate": 4.44943566238056e-05,
+      "loss": 0.4703,
       "step": 590
     },
     {
-      "epoch": 0.61,
-      "learning_rate": 3.950731479541743e-05,
-      "loss": 0.6698,
+      "epoch": 0.44,
+      "learning_rate": 4.431268932009865e-05,
+      "loss": 0.464,
       "step": 600
     },
     {
-      "epoch": 0.62,
-      "learning_rate": 3.918146159895882e-05,
-      "loss": 0.6549,
+      "epoch": 0.44,
+      "eval_loss": 0.46821752190589905,
+      "eval_runtime": 69.9505,
+      "eval_samples_per_second": 15.754,
+      "eval_steps_per_second": 1.587,
+      "step": 600
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 4.412845702454024e-05,
+      "loss": 0.4739,
       "step": 610
     },
     {
-      "epoch": 0.63,
-      "learning_rate": 3.8852016560078605e-05,
-      "loss": 0.6516,
+      "epoch": 0.45,
+      "learning_rate": 4.394168420572349e-05,
+      "loss": 0.4529,
       "step": 620
     },
     {
-      "epoch": 0.64,
-      "learning_rate": 3.851906311972943e-05,
-      "loss": 0.6629,
+      "epoch": 0.46,
+      "learning_rate": 4.3752395669658086e-05,
+      "loss": 0.4433,
       "step": 630
     },
     {
-      "epoch": 0.65,
-      "learning_rate": 3.821647502051616e-05,
-      "loss": 0.6764,
+      "epoch": 0.47,
+      "learning_rate": 4.356061655647571e-05,
+      "loss": 0.459,
       "step": 640
     },
     {
-      "epoch": 0.66,
-      "learning_rate": 3.787708866250794e-05,
-      "loss": 0.6415,
+      "epoch": 0.48,
+      "learning_rate": 4.336637233709107e-05,
+      "loss": 0.4877,
       "step": 650
     },
     {
-      "epoch": 0.67,
-      "learning_rate": 3.7534440830144466e-05,
-      "loss": 0.6463,
+      "epoch": 0.48,
+      "learning_rate": 4.316968880981903e-05,
+      "loss": 0.4676,
       "step": 660
     },
     {
-      "epoch": 0.68,
-      "learning_rate": 3.71886183083464e-05,
-      "loss": 0.6508,
+      "epoch": 0.49,
+      "learning_rate": 4.2970592096948236e-05,
+      "loss": 0.4782,
       "step": 670
     },
     {
-      "epoch": 0.69,
-      "learning_rate": 3.683970868611123e-05,
-      "loss": 0.6411,
+      "epoch": 0.5,
+      "learning_rate": 4.276910864127168e-05,
+      "loss": 0.4778,
       "step": 680
     },
     {
-      "epoch": 0.7,
-      "learning_rate": 3.648780033432891e-05,
-      "loss": 0.6266,
+      "epoch": 0.51,
+      "learning_rate": 4.25652652025748e-05,
+      "loss": 0.454,
       "step": 690
     },
     {
-      "epoch": 0.71,
-      "learning_rate": 3.613298238339955e-05,
-      "loss": 0.6409,
+      "epoch": 0.51,
+      "learning_rate": 4.235908885408133e-05,
+      "loss": 0.4399,
       "step": 700
     },
     {
-      "epoch": 0.72,
-      "learning_rate": 3.5775344700658705e-05,
-      "loss": 0.6594,
+      "epoch": 0.52,
+      "learning_rate": 4.215060697885767e-05,
+      "loss": 0.5106,
       "step": 710
     },
     {
-      "epoch": 0.73,
-      "learning_rate": 3.5414977867616006e-05,
-      "loss": 0.6427,
+      "epoch": 0.53,
+      "learning_rate": 4.1939847266176e-05,
+      "loss": 0.4683,
       "step": 720
     },
     {
-      "epoch": 0.74,
-      "learning_rate": 3.505197315701292e-05,
-      "loss": 0.6462,
+      "epoch": 0.54,
+      "learning_rate": 4.172683770783677e-05,
+      "loss": 0.4896,
       "step": 730
     },
     {
-      "epoch": 0.75,
-      "learning_rate": 3.468642250970547e-05,
-      "loss": 0.6277,
+      "epoch": 0.54,
+      "learning_rate": 4.1511606594451016e-05,
+      "loss": 0.4633,
       "step": 740
     },
     {
-      "epoch": 0.76,
-      "learning_rate": 3.431841851137764e-05,
-      "loss": 0.6551,
+      "epoch": 0.55,
+      "learning_rate": 4.1294182511682946e-05,
+      "loss": 0.4486,
       "step": 750
     },
     {
-      "epoch": 0.77,
-      "learning_rate": 3.394805436909157e-05,
-      "loss": 0.6402,
+      "epoch": 0.56,
+      "learning_rate": 4.1074594336453384e-05,
+      "loss": 0.4706,
       "step": 760
     },
     {
-      "epoch": 0.78,
-      "learning_rate": 3.357542388768033e-05,
-      "loss": 0.6515,
+      "epoch": 0.56,
+      "learning_rate": 4.085287123310455e-05,
+      "loss": 0.4797,
       "step": 770
     },
     {
-      "epoch": 0.79,
-      "learning_rate": 3.3200621445989226e-05,
-      "loss": 0.6489,
+      "epoch": 0.57,
+      "learning_rate": 4.062904264952657e-05,
+      "loss": 0.4251,
       "step": 780
     },
     {
-      "epoch": 0.8,
-      "learning_rate": 3.282374197297185e-05,
-      "loss": 0.6568,
+      "epoch": 0.58,
+      "learning_rate": 4.0403138313246435e-05,
+      "loss": 0.4687,
       "step": 790
     },
     {
-      "epoch": 0.81,
-      "learning_rate": 3.2444880923646674e-05,
-      "loss": 0.615,
+      "epoch": 0.59,
+      "learning_rate": 4.017518822747976e-05,
+      "loss": 0.48,
       "step": 800
     },
     {
-      "epoch": 0.82,
-      "learning_rate": 3.20641342549205e-05,
-      "loss": 0.6447,
+      "epoch": 0.59,
+      "eval_loss": 0.4554018974304199,
+      "eval_runtime": 70.2392,
+      "eval_samples_per_second": 15.689,
+      "eval_steps_per_second": 1.58,
+      "step": 800
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 3.994522266714594e-05,
+      "loss": 0.4662,
       "step": 810
     },
     {
-      "epoch": 0.83,
-      "learning_rate": 3.168159840128472e-05,
-      "loss": 0.6159,
+      "epoch": 0.6,
+      "learning_rate": 3.9713272174847246e-05,
+      "loss": 0.4837,
       "step": 820
     },
     {
-      "epoch": 0.84,
-      "learning_rate": 3.129737025039068e-05,
-      "loss": 0.6347,
+      "epoch": 0.61,
+      "learning_rate": 3.947936755681229e-05,
+      "loss": 0.5014,
       "step": 830
     },
     {
-      "epoch": 0.85,
-      "learning_rate": 3.091154711851022e-05,
-      "loss": 0.6361,
+      "epoch": 0.62,
+      "learning_rate": 3.92435398788046e-05,
+      "loss": 0.4764,
       "step": 840
     },
     {
-      "epoch": 0.86,
-      "learning_rate": 3.052422672588765e-05,
-      "loss": 0.6504,
+      "epoch": 0.62,
+      "learning_rate": 3.9005820461996604e-05,
+      "loss": 0.4587,
       "step": 850
     },
     {
-      "epoch": 0.87,
-      "learning_rate": 3.013550717198948e-05,
-      "loss": 0.6467,
+      "epoch": 0.63,
+      "learning_rate": 3.876624087880979e-05,
+      "loss": 0.484,
       "step": 860
     },
     {
-      "epoch": 0.88,
-      "learning_rate": 2.9745486910657993e-05,
-      "loss": 0.6364,
+      "epoch": 0.64,
+      "learning_rate": 3.85248329487214e-05,
+      "loss": 0.4285,
       "step": 870
     },
     {
-      "epoch": 0.89,
-      "learning_rate": 2.9354264725175185e-05,
-      "loss": 0.6361,
+      "epoch": 0.65,
+      "learning_rate": 3.828162873403843e-05,
+      "loss": 0.4742,
       "step": 880
     },
     {
-      "epoch": 0.9,
-      "learning_rate": 2.8961939703243122e-05,
-      "loss": 0.6441,
+      "epoch": 0.65,
+      "learning_rate": 3.803666053563926e-05,
+      "loss": 0.4533,
       "step": 890
     },
     {
-      "epoch": 0.91,
-      "learning_rate": 2.856861121188735e-05,
-      "loss": 0.6404,
+      "epoch": 0.66,
+      "learning_rate": 3.778996088868365e-05,
+      "loss": 0.4395,
       "step": 900
     },
     {
-      "epoch": 0.92,
-      "learning_rate": 2.8174378872289446e-05,
-      "loss": 0.6307,
+      "epoch": 0.67,
+      "learning_rate": 3.754156255829167e-05,
+      "loss": 0.4519,
       "step": 910
     },
     {
-      "epoch": 0.93,
-      "learning_rate": 2.777934253455522e-05,
-      "loss": 0.6484,
+      "epoch": 0.67,
+      "learning_rate": 3.7291498535191996e-05,
+      "loss": 0.4684,
       "step": 920
     },
     {
-      "epoch": 0.94,
-      "learning_rate": 2.7383602252424985e-05,
-      "loss": 0.6237,
+      "epoch": 0.68,
+      "learning_rate": 3.703980203134029e-05,
+      "loss": 0.4259,
       "step": 930
     },
     {
-      "epoch": 0.95,
-      "learning_rate": 2.6987258257932175e-05,
-      "loss": 0.6161,
+      "epoch": 0.69,
+      "learning_rate": 3.678650647550822e-05,
+      "loss": 0.4576,
       "step": 940
     },
     {
-      "epoch": 0.96,
-      "learning_rate": 2.6590410936016895e-05,
-      "loss": 0.6381,
+      "epoch": 0.7,
+      "learning_rate": 3.6531645508843636e-05,
+      "loss": 0.446,
       "step": 950
     },
     {
-      "epoch": 0.97,
-      "learning_rate": 2.619316079910063e-05,
-      "loss": 0.6366,
+      "epoch": 0.7,
+      "learning_rate": 3.6275252980402544e-05,
+      "loss": 0.4439,
       "step": 960
     },
     {
-      "epoch": 0.98,
-      "learning_rate": 2.5795608461628802e-05,
-      "loss": 0.6202,
+      "epoch": 0.71,
+      "learning_rate": 3.601736294265354e-05,
+      "loss": 0.476,
       "step": 970
     },
     {
-      "epoch": 0.99,
-      "learning_rate": 2.5397854614587334e-05,
-      "loss": 0.6334,
+      "epoch": 0.72,
+      "learning_rate": 3.5758009646955115e-05,
+      "loss": 0.4153,
       "step": 980
     },
     {
-      "epoch": 1.0,
-      "learning_rate": 2.5e-05,
-      "loss": 0.5954,
+      "epoch": 0.73,
+      "learning_rate": 3.5497227539006614e-05,
+      "loss": 0.4444,
       "step": 990
     },
     {
-      "epoch": 1.01,
-      "learning_rate": 2.460214538541267e-05,
-      "loss": 0.4963,
+      "epoch": 0.73,
+      "learning_rate": 3.523505125427341e-05,
+      "loss": 0.4395,
       "step": 1000
     },
     {
-      "epoch": 1.02,
-      "learning_rate": 2.4204391538371207e-05,
-      "loss": 0.487,
+      "epoch": 0.73,
+      "eval_loss": 0.44357746839523315,
+      "eval_runtime": 70.1513,
+      "eval_samples_per_second": 15.709,
+      "eval_steps_per_second": 1.582,
+      "step": 1000
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 3.497151561338678e-05,
+      "loss": 0.4387,
       "step": 1010
     },
     {
-      "epoch": 1.03,
-      "learning_rate": 2.3806839200899377e-05,
-      "loss": 0.489,
+      "epoch": 0.75,
+      "learning_rate": 3.470665561751928e-05,
+      "loss": 0.4559,
       "step": 1020
     },
     {
-      "epoch": 1.04,
-      "learning_rate": 2.3409589063983117e-05,
-      "loss": 0.4805,
+      "epoch": 0.76,
+      "learning_rate": 3.444050644373611e-05,
+      "loss": 0.4349,
       "step": 1030
     },
     {
-      "epoch": 1.05,
-      "learning_rate": 2.3012741742067838e-05,
-      "loss": 0.4907,
+      "epoch": 0.76,
+      "learning_rate": 3.417310344032309e-05,
+      "loss": 0.4661,
       "step": 1040
     },
     {
-      "epoch": 1.06,
-      "learning_rate": 2.261639774757503e-05,
-      "loss": 0.4719,
+      "epoch": 0.77,
+      "learning_rate": 3.390448212209191e-05,
+      "loss": 0.4945,
       "step": 1050
     },
     {
-      "epoch": 1.07,
-      "learning_rate": 2.2220657465444782e-05,
-      "loss": 0.4914,
+      "epoch": 0.78,
+      "learning_rate": 3.3634678165663325e-05,
+      "loss": 0.4399,
       "step": 1060
     },
     {
-      "epoch": 1.08,
-      "learning_rate": 2.182562112771056e-05,
-      "loss": 0.4775,
+      "epoch": 0.78,
+      "learning_rate": 3.336372740472877e-05,
+      "loss": 0.4399,
       "step": 1070
     },
     {
-      "epoch": 1.09,
-      "learning_rate": 2.143138878811265e-05,
-      "loss": 0.4935,
+      "epoch": 0.79,
+      "learning_rate": 3.309166582529114e-05,
+      "loss": 0.4251,
       "step": 1080
     },
     {
-      "epoch": 1.1,
-      "learning_rate": 2.1038060296756883e-05,
-      "loss": 0.5082,
+      "epoch": 0.8,
+      "learning_rate": 3.281852956088537e-05,
+      "loss": 0.4385,
       "step": 1090
     },
     {
-      "epoch": 1.11,
-      "learning_rate": 2.064573527482482e-05,
-      "loss": 0.4868,
+      "epoch": 0.81,
+      "learning_rate": 3.254435488777941e-05,
+      "loss": 0.4566,
       "step": 1100
     },
     {
-      "epoch": 1.12,
-      "learning_rate": 2.025451308934201e-05,
-      "loss": 0.4988,
+      "epoch": 0.81,
+      "learning_rate": 3.226917822015623e-05,
+      "loss": 0.4352,
       "step": 1110
     },
     {
-      "epoch": 1.13,
-      "learning_rate": 1.9864492828010526e-05,
-      "loss": 0.4653,
+      "epoch": 0.82,
+      "learning_rate": 3.199303610527749e-05,
+      "loss": 0.4617,
       "step": 1120
     },
     {
-      "epoch": 1.14,
-      "learning_rate": 1.9475773274112354e-05,
-      "loss": 0.4915,
+      "epoch": 0.83,
+      "learning_rate": 3.1715965218629595e-05,
+      "loss": 0.476,
       "step": 1130
     },
     {
-      "epoch": 1.16,
-      "learning_rate": 1.9088452881489787e-05,
-      "loss": 0.4763,
+      "epoch": 0.84,
+      "learning_rate": 3.143800235905268e-05,
+      "loss": 0.4322,
       "step": 1140
     },
     {
-      "epoch": 1.17,
-      "learning_rate": 1.8702629749609324e-05,
-      "loss": 0.4807,
+      "epoch": 0.84,
+      "learning_rate": 3.115918444385315e-05,
+      "loss": 0.4736,
       "step": 1150
     },
     {
-      "epoch": 1.18,
-      "learning_rate": 1.8318401598715284e-05,
-      "loss": 0.4653,
+      "epoch": 0.85,
+      "learning_rate": 3.0879548503900665e-05,
+      "loss": 0.4322,
       "step": 1160
     },
     {
-      "epoch": 1.19,
-      "learning_rate": 1.793586574507951e-05,
-      "loss": 0.4778,
+      "epoch": 0.86,
+      "learning_rate": 3.0599131678709836e-05,
+      "loss": 0.4152,
       "step": 1170
     },
     {
-      "epoch": 1.2,
-      "learning_rate": 1.7555119076353338e-05,
-      "loss": 0.4839,
+      "epoch": 0.87,
+      "learning_rate": 3.031797121150764e-05,
+      "loss": 0.4187,
       "step": 1180
     },
     {
-      "epoch": 1.21,
-      "learning_rate": 1.7176258027028152e-05,
-      "loss": 0.4718,
+      "epoch": 0.87,
+      "learning_rate": 3.0036104444286954e-05,
+      "loss": 0.4331,
       "step": 1190
     },
     {
-      "epoch": 1.22,
-      "learning_rate": 1.6799378554010773e-05,
-      "loss": 0.4793,
+      "epoch": 0.88,
+      "learning_rate": 2.9753568812847065e-05,
+      "loss": 0.45,
       "step": 1200
     },
     {
-      "epoch": 1.23,
-      "learning_rate": 1.6424576112319672e-05,
-      "loss": 0.4825,
+      "epoch": 0.88,
+      "eval_loss": 0.43782830238342285,
+      "eval_runtime": 70.0939,
+      "eval_samples_per_second": 15.722,
+      "eval_steps_per_second": 1.584,
+      "step": 1200
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 2.9470401841821686e-05,
+      "loss": 0.4457,
       "step": 1210
     },
     {
-      "epoch": 1.24,
-      "learning_rate": 1.6051945630908426e-05,
-      "loss": 0.4857,
+      "epoch": 0.89,
+      "learning_rate": 2.9186641139695108e-05,
+      "loss": 0.4667,
       "step": 1220
     },
     {
-      "epoch": 1.25,
-      "learning_rate": 1.5681581488622367e-05,
-      "loss": 0.4802,
+      "epoch": 0.9,
+      "learning_rate": 2.8902324393807333e-05,
+      "loss": 0.4516,
       "step": 1230
     },
     {
-      "epoch": 1.26,
-      "learning_rate": 1.5313577490294538e-05,
-      "loss": 0.4812,
+      "epoch": 0.91,
+      "learning_rate": 2.861748936534867e-05,
+      "loss": 0.4695,
       "step": 1240
     },
     {
-      "epoch": 1.27,
-      "learning_rate": 1.4948026842987084e-05,
-      "loss": 0.4682,
+      "epoch": 0.92,
+      "learning_rate": 2.8332173884344477e-05,
+      "loss": 0.45,
       "step": 1250
     },
     {
-      "epoch": 1.28,
-      "learning_rate": 1.4585022132384008e-05,
-      "loss": 0.4974,
+      "epoch": 0.92,
+      "learning_rate": 2.8046415844630857e-05,
+      "loss": 0.4246,
       "step": 1260
     },
     {
-      "epoch": 1.29,
-      "learning_rate": 1.4224655299341304e-05,
-      "loss": 0.4737,
+      "epoch": 0.93,
+      "learning_rate": 2.7760253198821822e-05,
+      "loss": 0.4449,
       "step": 1270
     },
     {
-      "epoch": 1.3,
-      "learning_rate": 1.3867017616600456e-05,
-      "loss": 0.4877,
+      "epoch": 0.94,
+      "learning_rate": 2.7473723953268687e-05,
+      "loss": 0.4217,
       "step": 1280
     },
     {
-      "epoch": 1.31,
-      "learning_rate": 1.3512199665671094e-05,
-      "loss": 0.4753,
+      "epoch": 0.95,
+      "learning_rate": 2.7186866163012232e-05,
+      "loss": 0.4495,
       "step": 1290
     },
     {
-      "epoch": 1.32,
-      "learning_rate": 1.316029131388878e-05,
-      "loss": 0.4638,
+      "epoch": 0.95,
+      "learning_rate": 2.6899717926728535e-05,
+      "loss": 0.4523,
       "step": 1300
     },
     {
-      "epoch": 1.33,
-      "learning_rate": 1.2811381691653607e-05,
-      "loss": 0.4626,
+      "epoch": 0.96,
+      "learning_rate": 2.6612317381668915e-05,
+      "loss": 0.4523,
       "step": 1310
     },
     {
-      "epoch": 1.34,
-      "learning_rate": 1.2465559169855535e-05,
-      "loss": 0.4786,
+      "epoch": 0.97,
+      "learning_rate": 2.632470269859478e-05,
+      "loss": 0.45,
       "step": 1320
     },
     {
-      "epoch": 1.35,
-      "learning_rate": 1.212291133749206e-05,
-      "loss": 0.4717,
+      "epoch": 0.98,
+      "learning_rate": 2.603691207670803e-05,
+      "loss": 0.4371,
       "step": 1330
     },
     {
-      "epoch": 1.36,
-      "learning_rate": 1.178352497948384e-05,
-      "loss": 0.4803,
+      "epoch": 0.98,
+      "learning_rate": 2.5748983738577653e-05,
+      "loss": 0.4221,
       "step": 1340
     },
     {
-      "epoch": 1.37,
-      "learning_rate": 1.1447486054694112e-05,
-      "loss": 0.4803,
+      "epoch": 0.99,
+      "learning_rate": 2.5460955925063268e-05,
+      "loss": 0.4565,
       "step": 1350
     },
     {
-      "epoch": 1.38,
-      "learning_rate": 1.1114879674157233e-05,
-      "loss": 0.4739,
+      "epoch": 1.0,
+      "learning_rate": 2.5172866890236203e-05,
+      "loss": 0.3892,
       "step": 1360
     },
     {
-      "epoch": 1.39,
-      "learning_rate": 1.0785790079522001e-05,
-      "loss": 0.471,
+      "epoch": 1.0,
+      "learning_rate": 2.48847548962988e-05,
+      "loss": 0.3973,
       "step": 1370
     },
     {
-      "epoch": 1.4,
-      "learning_rate": 1.046030062171512e-05,
-      "loss": 0.4799,
+      "epoch": 1.01,
+      "learning_rate": 2.4596658208502713e-05,
+      "loss": 0.4151,
       "step": 1380
     },
     {
-      "epoch": 1.41,
-      "learning_rate": 1.0138493739830352e-05,
-      "loss": 0.4689,
+      "epoch": 1.02,
+      "learning_rate": 2.4308615090066735e-05,
+      "loss": 0.3839,
       "step": 1390
     },
     {
-      "epoch": 1.42,
-      "learning_rate": 9.820450940248544e-06,
-      "loss": 0.4599,
+      "epoch": 1.03,
+      "learning_rate": 2.4020663797094864e-05,
+      "loss": 0.4062,
       "step": 1400
     },
     {
-      "epoch": 1.43,
-      "learning_rate": 9.506252775993882e-06,
-      "loss": 0.5019,
+      "epoch": 1.03,
+      "eval_loss": 0.43284282088279724,
+      "eval_runtime": 70.1658,
+      "eval_samples_per_second": 15.706,
+      "eval_steps_per_second": 1.582,
+      "step": 1400
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 2.373284257349544e-05,
+      "loss": 0.4003,
       "step": 1410
     },
     {
-      "epoch": 1.44,
-      "learning_rate": 9.195978826331697e-06,
-      "loss": 0.4764,
+      "epoch": 1.04,
+      "learning_rate": 2.3445189645901806e-05,
+      "loss": 0.4447,
       "step": 1420
     },
     {
-      "epoch": 1.45,
-      "learning_rate": 8.889707676612791e-06,
-      "loss": 0.4579,
+      "epoch": 1.05,
+      "learning_rate": 2.3157743218595247e-05,
+      "loss": 0.4565,
       "step": 1430
     },
     {
-      "epoch": 1.46,
-      "learning_rate": 8.587516898369589e-06,
-      "loss": 0.4592,
+      "epoch": 1.06,
+      "learning_rate": 2.287054146843097e-05,
+      "loss": 0.4265,
       "step": 1440
     },
     {
-      "epoch": 1.47,
-      "learning_rate": 8.289483029668972e-06,
-      "loss": 0.4861,
+      "epoch": 1.06,
+      "learning_rate": 2.2583622539767668e-05,
+      "loss": 0.4056,
       "step": 1450
     },
     {
-      "epoch": 1.48,
-      "learning_rate": 7.99568155572701e-06,
-      "loss": 0.4753,
+      "epoch": 1.07,
+      "learning_rate": 2.2297024539401463e-05,
+      "loss": 0.4074,
       "step": 1460
     },
     {
-      "epoch": 1.49,
-      "learning_rate": 7.706186889790209e-06,
-      "loss": 0.4929,
+      "epoch": 1.08,
+      "learning_rate": 2.2010785531504716e-05,
+      "loss": 0.4281,
       "step": 1470
     },
     {
-      "epoch": 1.5,
-      "learning_rate": 7.421072354288302e-06,
-      "loss": 0.4594,
+      "epoch": 1.09,
+      "learning_rate": 2.172494353257066e-05,
+      "loss": 0.4663,
       "step": 1480
     },
     {
-      "epoch": 1.51,
-      "learning_rate": 7.140410162263414e-06,
-      "loss": 0.4912,
+      "epoch": 1.09,
+      "learning_rate": 2.1439536506364274e-05,
+      "loss": 0.4281,
       "step": 1490
     },
     {
-      "epoch": 1.52,
-      "learning_rate": 6.86427139908008e-06,
-      "loss": 0.4681,
+      "epoch": 1.1,
+      "learning_rate": 2.1154602358880122e-05,
+      "loss": 0.4111,
       "step": 1500
     },
     {
-      "epoch": 1.53,
-      "learning_rate": 6.5927260044209655e-06,
-      "loss": 0.4816,
+      "epoch": 1.11,
+      "learning_rate": 2.0870178933307948e-05,
+      "loss": 0.4187,
       "step": 1510
     },
     {
-      "epoch": 1.54,
-      "learning_rate": 6.3258427545727e-06,
-      "loss": 0.4723,
+      "epoch": 1.11,
+      "learning_rate": 2.0586304005006585e-05,
+      "loss": 0.3964,
       "step": 1520
     },
     {
-      "epoch": 1.55,
-      "learning_rate": 6.063689245006443e-06,
-      "loss": 0.4856,
+      "epoch": 1.12,
+      "learning_rate": 2.030301527648684e-05,
+      "loss": 0.4331,
       "step": 1530
     },
     {
-      "epoch": 1.56,
-      "learning_rate": 5.806331873257462e-06,
-      "loss": 0.4829,
+      "epoch": 1.13,
+      "learning_rate": 2.0020350372404102e-05,
+      "loss": 0.4432,
       "step": 1540
     },
     {
-      "epoch": 1.57,
-      "learning_rate": 5.553835822108152e-06,
-      "loss": 0.4741,
+      "epoch": 1.14,
+      "learning_rate": 1.9738346834561254e-05,
+      "loss": 0.4056,
       "step": 1550
     },
     {
-      "epoch": 1.58,
-      "learning_rate": 5.306265043078693e-06,
-      "loss": 0.4654,
+      "epoch": 1.14,
+      "learning_rate": 1.945704211692262e-05,
+      "loss": 0.4354,
       "step": 1560
     },
     {
-      "epoch": 1.59,
-      "learning_rate": 5.0636822402296165e-06,
-      "loss": 0.4668,
+      "epoch": 1.15,
+      "learning_rate": 1.9176473580639538e-05,
+      "loss": 0.4309,
       "step": 1570
     },
     {
-      "epoch": 1.6,
-      "learning_rate": 4.826148854280277e-06,
-      "loss": 0.4723,
+      "epoch": 1.16,
+      "learning_rate": 1.8896678489088304e-05,
+      "loss": 0.4058,
       "step": 1580
     },
     {
-      "epoch": 1.61,
-      "learning_rate": 4.593725047047293e-06,
-      "loss": 0.4639,
+      "epoch": 1.17,
+      "learning_rate": 1.8617694002921064e-05,
+      "loss": 0.4319,
       "step": 1590
     },
     {
-      "epoch": 1.62,
-      "learning_rate": 4.3664696862069505e-06,
-      "loss": 0.4777,
+      "epoch": 1.17,
+      "learning_rate": 1.8339557175130383e-05,
+      "loss": 0.4267,
       "step": 1600
     },
     {
-      "epoch": 1.63,
-      "learning_rate": 4.144440330385347e-06,
-      "loss": 0.4546,
+      "epoch": 1.17,
+      "eval_loss": 0.42894992232322693,
+      "eval_runtime": 70.251,
+      "eval_samples_per_second": 15.687,
+      "eval_steps_per_second": 1.58,
+      "step": 1600
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.8062304946128073e-05,
+      "loss": 0.3921,
       "step": 1610
     },
     {
-      "epoch": 1.64,
-      "learning_rate": 3.927693214580075e-06,
-      "loss": 0.4543,
+      "epoch": 1.19,
+      "learning_rate": 1.7785974138839018e-05,
+      "loss": 0.4206,
       "step": 1620
     },
     {
-      "epoch": 1.65,
-      "learning_rate": 3.71628323591722e-06,
-      "loss": 0.4543,
+      "epoch": 1.2,
+      "learning_rate": 1.7510601453810594e-05,
+      "loss": 0.4356,
       "step": 1630
     },
     {
-      "epoch": 1.66,
-      "learning_rate": 3.5102639397471214e-06,
-      "loss": 0.4659,
+      "epoch": 1.2,
+      "learning_rate": 1.723622346433828e-05,
+      "loss": 0.4167,
       "step": 1640
     },
     {
-      "epoch": 1.67,
-      "learning_rate": 3.3096875060825845e-06,
-      "loss": 0.485,
+      "epoch": 1.21,
+      "learning_rate": 1.6962876611608262e-05,
+      "loss": 0.4233,
       "step": 1650
     },
     {
-      "epoch": 1.68,
-      "learning_rate": 3.11460473638282e-06,
-      "loss": 0.4768,
+      "epoch": 1.22,
+      "learning_rate": 1.6690597199857523e-05,
+      "loss": 0.4176,
       "step": 1660
     },
     {
-      "epoch": 1.69,
-      "learning_rate": 2.925065040686642e-06,
-      "loss": 0.4635,
+      "epoch": 1.22,
+      "learning_rate": 1.6419421391552142e-05,
+      "loss": 0.4241,
       "step": 1670
     },
     {
-      "epoch": 1.7,
-      "learning_rate": 2.741116425097995e-06,
-      "loss": 0.4681,
+      "epoch": 1.23,
+      "learning_rate": 1.6149385202584423e-05,
+      "loss": 0.4524,
       "step": 1680
     },
     {
-      "epoch": 1.71,
-      "learning_rate": 2.5628054796271063e-06,
-      "loss": 0.4492,
+      "epoch": 1.24,
+      "learning_rate": 1.5880524497489474e-05,
+      "loss": 0.4647,
       "step": 1690
     },
     {
-      "epoch": 1.72,
-      "learning_rate": 2.390177366390273e-06,
-      "loss": 0.4664,
+      "epoch": 1.25,
+      "learning_rate": 1.5612874984681923e-05,
+      "loss": 0.4036,
       "step": 1700
     },
     {
-      "epoch": 1.73,
-      "learning_rate": 2.22327580817136e-06,
-      "loss": 0.48,
+      "epoch": 1.25,
+      "learning_rate": 1.534647221171334e-05,
+      "loss": 0.4148,
       "step": 1710
     },
     {
-      "epoch": 1.74,
-      "learning_rate": 2.0621430773477947e-06,
-      "loss": 0.4616,
+      "epoch": 1.26,
+      "learning_rate": 1.5081351560551021e-05,
+      "loss": 0.408,
       "step": 1720
     },
     {
-      "epoch": 1.75,
-      "learning_rate": 1.906819985183908e-06,
-      "loss": 0.4854,
+      "epoch": 1.27,
+      "learning_rate": 1.4817548242878759e-05,
+      "loss": 0.457,
       "step": 1730
     },
     {
-      "epoch": 1.76,
-      "learning_rate": 1.7573458714944063e-06,
-      "loss": 0.4846,
+      "epoch": 1.28,
+      "learning_rate": 1.45550972954203e-05,
+      "loss": 0.4193,
       "step": 1740
     },
     {
-      "epoch": 1.77,
-      "learning_rate": 1.6137585946804674e-06,
-      "loss": 0.4552,
+      "epoch": 1.28,
+      "learning_rate": 1.4294033575285914e-05,
+      "loss": 0.4149,
       "step": 1750
     },
     {
-      "epoch": 1.78,
-      "learning_rate": 1.4760945221410638e-06,
-      "loss": 0.4615,
+      "epoch": 1.29,
+      "learning_rate": 1.4034391755342972e-05,
+      "loss": 0.4032,
       "step": 1760
     },
     {
-      "epoch": 1.79,
-      "learning_rate": 1.3443885210619428e-06,
-      "loss": 0.4735,
+      "epoch": 1.3,
+      "learning_rate": 1.3776206319610823e-05,
+      "loss": 0.4078,
       "step": 1770
     },
     {
-      "epoch": 1.8,
-      "learning_rate": 1.2186739495845477e-06,
-      "loss": 0.4705,
+      "epoch": 1.31,
+      "learning_rate": 1.3519511558680892e-05,
+      "loss": 0.4123,
       "step": 1780
     },
     {
-      "epoch": 1.81,
-      "learning_rate": 1.0989826483571552e-06,
-      "loss": 0.4653,
+      "epoch": 1.31,
+      "learning_rate": 1.3264341565162422e-05,
+      "loss": 0.4089,
       "step": 1790
     },
     {
-      "epoch": 1.82,
-      "learning_rate": 9.85344932470364e-07,
-      "loss": 0.4632,
+      "epoch": 1.32,
+      "learning_rate": 1.3010730229154445e-05,
+      "loss": 0.383,
       "step": 1800
     },
     {
-      "epoch": 1.83,
-      "learning_rate": 8.77789583778979e-07,
-      "loss": 0.481,
+      "epoch": 1.32,
+      "eval_loss": 0.4250344932079315,
+      "eval_runtime": 70.4814,
+      "eval_samples_per_second": 15.635,
+      "eval_steps_per_second": 1.575,
+      "step": 1800
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 1.2758711233744783e-05,
+      "loss": 0.3936,
       "step": 1810
     },
     {
-      "epoch": 1.84,
-      "learning_rate": 7.763438436122122e-07,
-      "loss": 0.4773,
+      "epoch": 1.33,
+      "learning_rate": 1.2508318050536421e-05,
+      "loss": 0.4127,
       "step": 1820
     },
     {
-      "epoch": 1.85,
-      "learning_rate": 6.810334058740736e-07,
-      "loss": 0.4791,
+      "epoch": 1.34,
+      "learning_rate": 1.2259583935202062e-05,
+      "loss": 0.4015,
       "step": 1830
     },
     {
-      "epoch": 1.86,
-      "learning_rate": 5.918824105356797e-07,
-      "loss": 0.4793,
+      "epoch": 1.35,
+      "learning_rate": 1.2012541923067244e-05,
+      "loss": 0.4203,
       "step": 1840
     },
     {
-      "epoch": 1.87,
-      "learning_rate": 5.08913437521169e-07,
-      "loss": 0.4647,
+      "epoch": 1.36,
+      "learning_rate": 1.176722482472286e-05,
+      "loss": 0.4367,
       "step": 1850
     },
     {
-      "epoch": 1.88,
-      "learning_rate": 4.3214750098869995e-07,
-      "loss": 0.4765,
+      "epoch": 1.36,
+      "learning_rate": 1.1523665221667398e-05,
+      "loss": 0.4233,
       "step": 1860
     },
     {
-      "epoch": 1.89,
-      "learning_rate": 3.616040440080432e-07,
-      "loss": 0.4787,
+      "epoch": 1.37,
+      "learning_rate": 1.1281895461979732e-05,
+      "loss": 0.4405,
       "step": 1870
     },
     {
-      "epoch": 1.9,
-      "learning_rate": 2.973009336361021e-07,
-      "loss": 0.4723,
+      "epoch": 1.38,
+      "learning_rate": 1.104194765602281e-05,
+      "loss": 0.4379,
       "step": 1880
     },
     {
-      "epoch": 1.91,
-      "learning_rate": 2.392544563915883e-07,
-      "loss": 0.453,
+      "epoch": 1.39,
+      "learning_rate": 1.0803853672178946e-05,
+      "loss": 0.4146,
       "step": 1890
     },
     {
-      "epoch": 1.93,
-      "learning_rate": 1.8747931413001795e-07,
-      "loss": 0.4561,
+      "epoch": 1.39,
+      "learning_rate": 1.0567645132617316e-05,
+      "loss": 0.4438,
       "step": 1900
     },
     {
-      "epoch": 1.94,
-      "learning_rate": 1.4198862032005488e-07,
-      "loss": 0.4612,
+      "epoch": 1.4,
+      "learning_rate": 1.0333353409094015e-05,
+      "loss": 0.3915,
       "step": 1910
     },
     {
-      "epoch": 1.95,
-      "learning_rate": 1.0279389672218365e-07,
-      "loss": 0.4774,
+      "epoch": 1.41,
+      "learning_rate": 1.0101009618785528e-05,
+      "loss": 0.4063,
       "step": 1920
     },
     {
-      "epoch": 1.96,
-      "learning_rate": 6.990507047049676e-08,
-      "loss": 0.4635,
+      "epoch": 1.42,
+      "learning_rate": 9.870644620155877e-06,
+      "loss": 0.3974,
       "step": 1930
     },
     {
-      "epoch": 1.97,
-      "learning_rate": 4.3330471558378213e-08,
-      "loss": 0.4761,
+      "epoch": 1.42,
+      "learning_rate": 9.642289008858244e-06,
+      "loss": 0.4015,
       "step": 1940
     },
     {
-      "epoch": 1.98,
-      "learning_rate": 2.3076830728713252e-08,
-      "loss": 0.4593,
+      "epoch": 1.43,
+      "learning_rate": 9.41597311367142e-06,
+      "loss": 0.4264,
       "step": 1950
     },
     {
-      "epoch": 1.99,
-      "learning_rate": 9.149277769132658e-09,
-      "loss": 0.4736,
+      "epoch": 1.44,
+      "learning_rate": 9.191726992471725e-06,
+      "loss": 0.4334,
       "step": 1960
     },
     {
-      "epoch": 2.0,
-      "learning_rate": 1.551340212760377e-09,
-      "loss": 0.4486,
+      "epoch": 1.44,
+      "learning_rate": 8.969580428240903e-06,
+      "loss": 0.4197,
       "step": 1970
     },
+    {
+      "epoch": 1.45,
+      "learning_rate": 8.74956292511056e-06,
+      "loss": 0.422,
+      "step": 1980
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 8.531703704443575e-06,
+      "loss": 0.4077,
+      "step": 1990
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 8.316031700953086e-06,
+      "loss": 0.4334,
+      "step": 2000
+    },
+    {
+      "epoch": 1.47,
+      "eval_loss": 0.4206145107746124,
+      "eval_runtime": 70.0533,
+      "eval_samples_per_second": 15.731,
+      "eval_steps_per_second": 1.585,
+      "step": 2000
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 8.102575558859612e-06,
+      "loss": 0.4171,
+      "step": 2010
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 7.891363628086671e-06,
+      "loss": 0.3957,
+      "step": 2020
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 7.682423960495517e-06,
+      "loss": 0.4174,
+      "step": 2030
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 7.475784306159478e-06,
+      "loss": 0.4187,
+      "step": 2040
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 7.271472109678379e-06,
+      "loss": 0.3775,
+      "step": 2050
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 7.0695145065334585e-06,
+      "loss": 0.4019,
+      "step": 2060
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 6.869938319483471e-06,
+      "loss": 0.4143,
+      "step": 2070
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 6.67277005500222e-06,
+      "loss": 0.3853,
+      "step": 2080
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 6.478035899758139e-06,
+      "loss": 0.4403,
+      "step": 2090
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 6.285761717136335e-06,
+      "loss": 0.3888,
+      "step": 2100
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 6.095973043803577e-06,
+      "loss": 0.4123,
+      "step": 2110
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 5.908695086316701e-06,
+      "loss": 0.4035,
+      "step": 2120
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 5.723952717774763e-06,
+      "loss": 0.4219,
+      "step": 2130
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 5.54177047451562e-06,
+      "loss": 0.4209,
+      "step": 2140
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 5.362172552857128e-06,
+      "loss": 0.4286,
+      "step": 2150
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 5.18518280588354e-06,
+      "loss": 0.4134,
+      "step": 2160
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 5.010824740277501e-06,
+      "loss": 0.4005,
+      "step": 2170
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 4.83912151319802e-06,
+      "loss": 0.3973,
+      "step": 2180
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 4.6700959292048875e-06,
+      "loss": 0.3959,
+      "step": 2190
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 4.503770437229918e-06,
+      "loss": 0.4017,
+      "step": 2200
+    },
+    {
+      "epoch": 1.61,
+      "eval_loss": 0.41970890760421753,
+      "eval_runtime": 69.9019,
+      "eval_samples_per_second": 15.765,
+      "eval_steps_per_second": 1.588,
+      "step": 2200
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 4.340167127595407e-06,
+      "loss": 0.4013,
+      "step": 2210
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 4.179307729080256e-06,
+      "loss": 0.4069,
+      "step": 2220
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 4.021213606034064e-06,
+      "loss": 0.3857,
+      "step": 2230
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 3.8659057555396645e-06,
+      "loss": 0.4262,
+      "step": 2240
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 3.713404804624418e-06,
+      "loss": 0.3953,
+      "step": 2250
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 3.5637310075206544e-06,
+      "loss": 0.4231,
+      "step": 2260
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 3.416904242975635e-06,
+      "loss": 0.407,
+      "step": 2270
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 3.2729440116113843e-06,
+      "loss": 0.4254,
+      "step": 2280
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 3.131869433334725e-06,
+      "loss": 0.4054,
+      "step": 2290
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 2.9936992447979068e-06,
+      "loss": 0.3801,
+      "step": 2300
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 2.8584517969101054e-06,
+      "loss": 0.4335,
+      "step": 2310
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 2.7261450524001807e-06,
+      "loss": 0.412,
+      "step": 2320
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 2.596796583430969e-06,
+      "loss": 0.4456,
+      "step": 2330
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 2.470423569265462e-06,
+      "loss": 0.3972,
+      "step": 2340
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 2.34704279398516e-06,
+      "loss": 0.3935,
+      "step": 2350
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 2.2266706442609226e-06,
+      "loss": 0.3974,
+      "step": 2360
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 2.109323107176578e-06,
+      "loss": 0.4157,
+      "step": 2370
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 1.9950157681056318e-06,
+      "loss": 0.4379,
+      "step": 2380
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 1.8837638086413063e-06,
+      "loss": 0.4035,
+      "step": 2390
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 1.7755820045802145e-06,
+      "loss": 0.4455,
+      "step": 2400
+    },
+    {
+      "epoch": 1.76,
+      "eval_loss": 0.41885045170783997,
+      "eval_runtime": 70.0957,
+      "eval_samples_per_second": 15.721,
+      "eval_steps_per_second": 1.584,
+      "step": 2400
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 1.6704847239599364e-06,
+      "loss": 0.4532,
+      "step": 2410
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 1.5684859251507394e-06,
+      "loss": 0.4213,
+      "step": 2420
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 1.4695991550017164e-06,
+      "loss": 0.3873,
+      "step": 2430
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 1.373837547041576e-06,
+      "loss": 0.4163,
+      "step": 2440
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 1.2812138197343392e-06,
+      "loss": 0.3946,
+      "step": 2450
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 1.1917402747901152e-06,
+      "loss": 0.41,
+      "step": 2460
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 1.105428795531327e-06,
+      "loss": 0.4126,
+      "step": 2470
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 1.0222908453143804e-06,
+      "loss": 0.4505,
+      "step": 2480
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 9.423374660072065e-07,
+      "loss": 0.4303,
+      "step": 2490
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 8.655792765227405e-07,
+      "loss": 0.4179,
+      "step": 2500
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 7.920264714085828e-07,
+      "loss": 0.4376,
+      "step": 2510
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 7.216888194930272e-07,
+      "loss": 0.4069,
+      "step": 2520
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 6.545756625876031e-07,
+      "loss": 0.3887,
+      "step": 2530
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 5.906959142463947e-07,
+      "loss": 0.3972,
+      "step": 2540
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 5.300580585821696e-07,
+      "loss": 0.4078,
+      "step": 2550
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 4.7267014913956463e-07,
+      "loss": 0.4467,
+      "step": 2560
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 4.1853980782549097e-07,
+      "loss": 0.4567,
+      "step": 2570
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 3.6767422389682173e-07,
+      "loss": 0.4494,
+      "step": 2580
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 3.2008015300555306e-07,
+      "loss": 0.4344,
+      "step": 2590
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 2.757639163015774e-07,
+      "loss": 0.4101,
+      "step": 2600
+    },
+    {
+      "epoch": 1.91,
+      "eval_loss": 0.41886425018310547,
+      "eval_runtime": 70.2179,
+      "eval_samples_per_second": 15.694,
+      "eval_steps_per_second": 1.581,
+      "step": 2600
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 2.347313995931466e-07,
+      "loss": 0.4058,
+      "step": 2610
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 1.9698805256513908e-07,
+      "loss": 0.414,
+      "step": 2620
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 1.6253888805527474e-07,
+      "loss": 0.4254,
+      "step": 2630
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 1.3138848138835313e-07,
+      "loss": 0.4017,
+      "step": 2640
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 1.0354096976856186e-07,
+      "loss": 0.4258,
+      "step": 2650
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 7.900005173002712e-08,
+      "loss": 0.4098,
+      "step": 2660
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 5.776898664557051e-08,
+      "loss": 0.42,
+      "step": 2670
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 3.985059429383875e-08,
+      "loss": 0.3886,
+      "step": 2680
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 2.5247254484794813e-08,
+      "loss": 0.4276,
+      "step": 2690
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 1.3960906743634706e-08,
+      "loss": 0.4006,
+      "step": 2700
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 5.993050053204607e-09,
+      "loss": 0.4243,
+      "step": 2710
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 1.3447426549129117e-09,
+      "loss": 0.4439,
+      "step": 2720
+    },
     {
       "epoch": 2.0,
-      "step": 1974,
-      "total_flos": 3.437359013145084e+19,
-      "train_loss": 0.5708327819994277,
-      "train_runtime": 105327.89,
-      "train_samples_per_second": 4.797,
-      "train_steps_per_second": 0.019
+      "step": 2726,
+      "total_flos": 1.1289170586615415e+19,
+      "train_loss": 0.46561298847898175,
+      "train_runtime": 35477.4141,
+      "train_samples_per_second": 6.149,
+      "train_steps_per_second": 0.077
     }
   ],
-  "max_steps": 1974,
+  "max_steps": 2726,
   "num_train_epochs": 2,
-  "total_flos": 3.437359013145084e+19,
+  "total_flos": 1.1289170586615415e+19,
   "trial_name": null,
   "trial_params": null
 }

{Baichuan-13B-Chat-full → Baichuan-13B-Chat-lora-Task}/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fe0b7bf81fc699ffb5b38f22a7bd9ad201bc6a51a7928e5d8ae8e14abe781142
-size 5752
+oid sha256:0037c3e69fa21891dd65687a3f79ad20a8abc0a75790c554b437378b35c59fe9
+size 4600

README.md

@@ -1,3 +1,150 @@
 ---
 license: apache-2.0
+language:
+- zh
+tags:
+- finance
 ---
+
+This repository contains the DISC-FinLLM, version of Baichuan-13B-Chat as the base model.
+
+<div align="center">
+  
+[Demo](https://law.fudan-disc.com) | [技术报告](https://arxiv.org/abs/2309.11325)
+</div>
+
+**Please note that due to the ongoing development of the project, the model weights in this repository may differ from those in our currently deployed demo.**
+
+
+DISC-LawLLM is a large language model specialized in Chinese legal domain, developed and open-sourced by [Data Intelligence and Social Computing Lab of Fudan University  (Fudan-DISC)](http://fudan-disc.com), to provide comprehensive intelligent legal services. The advtantages is:
+* **Legal Texts Generic Processing Capability** 
+* **Legal Thinking and Reasoning** 
+* **Legal knowledge Retrieval Capacity** 
+
+In addition, the contributions include:
+
+* **High-quality SFT datasets and effective training paradigms**
+* **Chinese legal LLMs evaluation framework**
+  
+Check our [HOME](https://github.com/FudanDISC/DISC-LawLLM) for more information. 
+
+# DISC-Law-SFT Dataset
+
+we construct a high-quality supervised fine-tuning dataset, DISC-Law-SFT with two subsets, namely DISC-Law-SFT-Pair and DISC-Law-SFT-Triplet.  Our dataset converge a range of legal tasks, including legal information extraction, judgment prediction, document summarization, and legal question answering, ensuring coverage of diverse scenarios.
+<img src="" alt="" width=""/>
+
+<table>
+  <tr>
+    <th>Dataset</th>
+    <th>Task/Source</th>
+    <th>Size</th>
+    <th>Scenario</th>
+  </tr>
+  <tr>
+    <td rowspan="10">DISC-LawLLM-SFT-Pair</td>
+    <td>Legal information extraction</td>
+    <td>32K</td>
+    <td rowspan="7">Legal professional assistant</td>
+  </tr>
+  <tr>
+    <td>Legal event detection</td>
+    <td>27K</td>
+  </tr>
+  <tr>
+    <td>Legal case classification</td>
+    <td>20K</td>
+  </tr>
+  <tr>
+    <td>Legal judgement prediction</td>
+    <td>11K</td>
+  </tr>
+  <tr>
+    <td>Legal case matching</td>
+    <td>8K</td>
+  </tr>
+  <tr>
+    <td>Legal text summarization</td>
+    <td>9K</td>
+  </tr>
+  <tr>
+    <td>Judicial public opinion summarization</td>
+    <td>6K</td>
+  </tr>
+  <tr>
+    <td>Legal question answering</td>
+    <td>93K</td>
+    <td>Legal consultation services</td>
+  </tr>
+  <tr>
+    <td>Legal reading comprehension</td>
+    <td>38K</td>
+    <td rowspan="2">Judicial examination assistant</td>
+  </tr>
+  <tr>
+    <td>Judicial examination</td>
+    <td>12K</td>
+  </tr>
+  <tr>
+    <td rowspan="2">DISC-LawLLM-SFT-Triple</td>
+    <td>Legal judgement prediction</td>
+    <td>16K</td>
+    <td>Legal professional assistant</td>
+  </tr>
+  <tr>
+    <td>Legal question answering</td>
+    <td>23K</td>
+    <td>Legal consultation services</td>
+  </tr>
+  <tr>
+    <td rowspan="2">General</td>
+    <td>Alpaca-GPT4</td>
+    <td>48K</td>
+    <td rowspan="2">General scenarios</td>
+  </tr>
+  <tr>
+    <td>Firefly</td>
+    <td>60K</td>
+  </tr>
+  <tr>
+    <td>Total</td>
+    <td colspan="3">403K</td>
+  </tr>
+</table>
+
+# Using through hugging face transformers
+
+```python
+>>>import torch
+>>>>>>from transformers import AutoModelForCausalLM, AutoTokenizer
+>>>from transformers.generation.utils import GenerationConfig
+>>>tokenizer = AutoTokenizer.from_pretrained("ShengbinYue/DISC-LawLLM", use_fast=False, trust_remote_code=True)
+>>>model = AutoModelForCausalLM.from_pretrained("ShengbinYue/DISC-LawLLM", device_map="auto", torch_dtype=torch.float16, trust_remote_code=True)
+>>>model.generation_config = GenerationConfig.from_pretrained("ShengbinYue/DISC-LawLLM")
+>>>messages = []
+>>>messages.append({"role": "user", "content": "生产销售假冒伪劣商品罪如何判刑？"})
+>>>response = model.chat(tokenizer, messages)
+>>>print(response)
+```
+
+# Disclaimer
+
+DISC-LawLLM comes with issues and limitations that current LLMs have yet to overcome. While it can provide Chinese legal services in many a wide variety of tasks and scenarios, the model should be used for reference purposes only and cannot replace professional lawyers and legal experts. We encourage users of DISC-LawLLM to evaluate the model critically. We do not take responsibility for any issues, risks, or adverse consequences that may arise from the use of DISC-LawLLM.
+
+# Citation
+
+If our work is helpful for your, please kindly cite our work as follows:
+
+```
+@misc{yue2023disclawllm,
+    title={DISC-LawLLM: Fine-tuning Large Language Models for Intelligent Legal Services}, 
+    author={Shengbin Yue and Wei Chen and Siyuan Wang and Bingxuan Li and Chenchen Shen and Shujun Liu and Yuxuan Zhou and Yao Xiao and Song Yun and Wei Lin and Xuanjing Huang and Zhongyu Wei},
+    year={2023},
+    eprint={2309.11325},
+    archivePrefix={arXiv},
+    primaryClass={cs.CL}
+}
+```
+
+# License
+
+The use of the source code in this repository complies with the Apache 2.0 License.