add

README.md

@@ -0,0 +1,20 @@
+---
+library_name: peft
+---
+## Training procedure
+
+
+The following `bitsandbytes` quantization config was used during training:
+- load_in_8bit: False
+- load_in_4bit: True
+- llm_int8_threshold: 6.0
+- llm_int8_skip_modules: None
+- llm_int8_enable_fp32_cpu_offload: False
+- llm_int8_has_fp16_weight: False
+- bnb_4bit_quant_type: nf4
+- bnb_4bit_use_double_quant: True
+- bnb_4bit_compute_dtype: float16
+### Framework versions
+
+
+- PEFT 0.4.0

adapter_config.json

@@ -0,0 +1,21 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": "../internlm-chat-20b",
+  "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": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM"
+}

adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8692759d0e4907942aecd47602d501e3c0d99b8481710db9efa4c347ae2bbecb
+size 39407885

all_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 2.0,
+    "train_loss": 1.4111853902105498,
+    "train_runtime": 51425.1791,
+    "train_samples_per_second": 2.684,
+    "train_steps_per_second": 0.042
+}

checkpoint-1000/README.md

@@ -0,0 +1,20 @@
+---
+library_name: peft
+---
+## Training procedure
+
+
+The following `bitsandbytes` quantization config was used during training:
+- load_in_8bit: False
+- load_in_4bit: True
+- llm_int8_threshold: 6.0
+- llm_int8_skip_modules: None
+- llm_int8_enable_fp32_cpu_offload: False
+- llm_int8_has_fp16_weight: False
+- bnb_4bit_quant_type: nf4
+- bnb_4bit_use_double_quant: True
+- bnb_4bit_compute_dtype: float16
+### Framework versions
+
+
+- PEFT 0.4.0

checkpoint-1000/adapter_config.json

@@ -0,0 +1,21 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": "../internlm-chat-20b",
+  "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": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM"
+}

checkpoint-1000/adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4998260bd526a9c3bee94abe2e54be3fec53559ce12bc659c21899480083d3e6
+size 39407885

checkpoint-1000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5fa4225c75d79ec4376d38b077a7cdff57bb875e08cc33c5d567919695684878
+size 78784709

checkpoint-1000/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eb09c0ab810491cec8dcbb357118e13006f5424ce1d217c048964a2052a80e9f
+size 18679

checkpoint-1000/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:addb671ff88a73231dd9b55b4517a019907c186cc25e81845c679685a9459723
+size 18679

checkpoint-1000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b3cf3e0db7ad9e9ba5dd577b3481b4ddbb58c5600860f207a1a04a3a4ee97a42
+size 627

checkpoint-1000/special_tokens_map.json

@@ -0,0 +1,9 @@
+{
+  "additional_special_tokens": [
+    "<eoa>"
+  ],
+  "bos_token": "<s>",
+  "eos_token": "</s>",
+  "pad_token": "</s>",
+  "unk_token": "<unk>"
+}

checkpoint-1000/tokenization_internlm.py

@@ -0,0 +1,242 @@
+# coding=utf-8
+# Copyright 2022 EleutherAI and the HuggingFace Inc. team. All rights reserved.
+#
+# This code is based on EleutherAI's GPT-NeoX library and the GPT-NeoX
+# and OPT implementations in this library. It has been modified from its
+# original forms to accommodate minor architectural differences compared
+# to GPT-NeoX and OPT used by the Meta AI team that trained the model.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+"""Tokenization classes for IntermLM."""
+import os
+from shutil import copyfile
+from typing import Any, Dict, List, Optional, Tuple
+
+import sentencepiece as spm
+
+from transformers.tokenization_utils import PreTrainedTokenizer
+from transformers.utils import logging
+
+
+logger = logging.get_logger(__name__)
+
+VOCAB_FILES_NAMES = {"vocab_file": "./tokenizer.model"}
+
+PRETRAINED_VOCAB_FILES_MAP = {}
+
+
+class InternLMTokenizer(PreTrainedTokenizer):
+    """
+    Construct a InternLM tokenizer. Based on byte-level Byte-Pair-Encoding.
+
+    Args:
+        vocab_file (`str`):
+            Path to the vocabulary file.
+    """
+
+    vocab_files_names = VOCAB_FILES_NAMES
+    pretrained_vocab_files_map = PRETRAINED_VOCAB_FILES_MAP
+    model_input_names = ["input_ids", "attention_mask"]
+    _auto_class = "AutoTokenizer"
+
+    def __init__(
+        self,
+        vocab_file,
+        unk_token="<unk>",
+        bos_token="<s>",
+        eos_token="</s>",
+        pad_token="</s>",
+        sp_model_kwargs: Optional[Dict[str, Any]] = None,
+        add_bos_token=True,
+        add_eos_token=False,
+        decode_with_prefix_space=False,
+        clean_up_tokenization_spaces=False,
+        **kwargs,
+    ):
+        self.sp_model_kwargs = {} if sp_model_kwargs is None else sp_model_kwargs
+        super().__init__(
+            bos_token=bos_token,
+            eos_token=eos_token,
+            unk_token=unk_token,
+            pad_token=pad_token,
+            clean_up_tokenization_spaces=clean_up_tokenization_spaces,
+            **kwargs,
+        )
+        self.vocab_file = vocab_file
+        self.add_bos_token = add_bos_token
+        self.add_eos_token = add_eos_token
+        self.decode_with_prefix_space = decode_with_prefix_space
+        self.sp_model = spm.SentencePieceProcessor(**self.sp_model_kwargs)
+        self.sp_model.Load(vocab_file)
+        self._no_prefix_space_tokens = None
+
+        """ Initialisation"""
+
+    @property
+    def no_prefix_space_tokens(self):
+        if self._no_prefix_space_tokens is None:
+            vocab = self.convert_ids_to_tokens(list(range(self.vocab_size)))
+            self._no_prefix_space_tokens = {i for i, tok in enumerate(vocab) if not tok.startswith("▁")}
+        return self._no_prefix_space_tokens
+
+    @property
+    def vocab_size(self):
+        """Returns vocab size"""
+        return self.sp_model.get_piece_size()
+
+    @property
+    def bos_token_id(self) -> Optional[int]:
+        return self.sp_model.bos_id()
+
+    @property
+    def eos_token_id(self) -> Optional[int]:
+        return self.sp_model.eos_id()
+
+    def get_vocab(self):
+        """Returns vocab as a dict"""
+        vocab = {self.convert_ids_to_tokens(i): i for i in range(self.vocab_size)}
+        vocab.update(self.added_tokens_encoder)
+        return vocab
+
+    def _tokenize(self, text):
+        """Returns a tokenized string."""
+        return self.sp_model.encode(text, out_type=str)
+
+    def _convert_token_to_id(self, token):
+        """Converts a token (str) in an id using the vocab."""
+        return self.sp_model.piece_to_id(token)
+
+    def _convert_id_to_token(self, index):
+        """Converts an index (integer) in a token (str) using the vocab."""
+        token = self.sp_model.IdToPiece(index)
+        return token
+
+    def _maybe_add_prefix_space(self, tokens, decoded):
+        if tokens and tokens[0] not in self.no_prefix_space_tokens:
+            return " " + decoded
+        else:
+            return decoded
+
+    def convert_tokens_to_string(self, tokens):
+        """Converts a sequence of tokens (string) in a single string."""
+        current_sub_tokens = []
+        out_string = ""
+        prev_is_special = False
+        for token in tokens:
+            # make sure that special tokens are not decoded using sentencepiece model
+            if token in self.all_special_tokens:
+                if not prev_is_special:
+                    out_string += " "
+                out_string += self.sp_model.decode(current_sub_tokens) + token
+                prev_is_special = True
+                current_sub_tokens = []
+            else:
+                current_sub_tokens.append(token)
+                prev_is_special = False
+        out_string += self.sp_model.decode(current_sub_tokens)
+        out_string = self.clean_up_tokenization(out_string)
+        out_string = self._maybe_add_prefix_space(tokens=tokens, decoded=out_string)
+        return out_string[1:]
+
+    def save_vocabulary(self, save_directory, filename_prefix: Optional[str] = None) -> Tuple[str]:
+        """
+        Save the vocabulary and special tokens file to a directory.
+
+        Args:
+            save_directory (`str`):
+                The directory in which to save the vocabulary.
+
+        Returns:
+            `Tuple(str)`: Paths to the files saved.
+        """
+        if not os.path.isdir(save_directory):
+            logger.error(f"Vocabulary path ({save_directory}) should be a directory")
+            return
+        out_vocab_file = os.path.join(
+            save_directory, (filename_prefix + "-" if filename_prefix else "") + VOCAB_FILES_NAMES["vocab_file"]
+        )
+
+        if os.path.abspath(self.vocab_file) != os.path.abspath(out_vocab_file) and os.path.isfile(self.vocab_file):
+            copyfile(self.vocab_file, out_vocab_file)
+        elif not os.path.isfile(self.vocab_file):
+            with open(out_vocab_file, "wb") as fi:
+                content_spiece_model = self.sp_model.serialized_model_proto()
+                fi.write(content_spiece_model)
+
+        return (out_vocab_file,)
+
+    def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=None):
+        if self.add_bos_token:
+            bos_token_ids = [self.bos_token_id]
+        else:
+            bos_token_ids = []
+
+        output = bos_token_ids + token_ids_0
+
+        if token_ids_1 is not None:
+            output = output + token_ids_1
+
+        if self.add_eos_token:
+            output = output + [self.eos_token_id]
+
+        return output
+
+    def get_special_tokens_mask(
+        self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None, already_has_special_tokens: bool = False
+    ) -> List[int]:
+        """
+        Retrieve sequence ids from a token list that has no special tokens added. This method is called when adding
+        special tokens using the tokenizer `prepare_for_model` method.
+
+        Args:
+            token_ids_0 (`List[int]`):
+                List of IDs.
+            token_ids_1 (`List[int]`, *optional*):
+                Optional second list of IDs for sequence pairs.
+            already_has_special_tokens (`bool`, *optional*, defaults to `False`):
+                Whether or not the token list is already formatted with special tokens for the model.
+
+        Returns:
+            `List[int]`: A list of integers in the range [0, 1]: 1 for a special token, 0 for a sequence token.
+        """
+        if already_has_special_tokens:
+            return super().get_special_tokens_mask(
+                token_ids_0=token_ids_0, token_ids_1=token_ids_1, already_has_special_tokens=True
+            )
+
+        if token_ids_1 is None:
+            return [1] + ([0] * len(token_ids_0)) + [1]
+        return [1] + ([0] * len(token_ids_0)) + [1, 1] + ([0] * len(token_ids_1)) + [1]
+
+    def create_token_type_ids_from_sequences(
+        self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None
+    ) -> List[int]:
+        """
+        Create a mask from the two sequences passed to be used in a sequence-pair classification task. T5 does not make
+        use of token type ids, therefore a list of zeros is returned.
+
+        Args:
+            token_ids_0 (`List[int]`):
+                List of IDs.
+            token_ids_1 (`List[int]`, *optional*):
+                Optional second list of IDs for sequence pairs.
+
+        Returns:
+            `List[int]`: List of zeros.
+        """
+        eos = [self.eos_token_id]
+
+        if token_ids_1 is None:
+            return len(token_ids_0 + eos) * [0]
+        return len(token_ids_0 + eos + token_ids_1 + eos) * [0]

checkpoint-1000/tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aab622d98c98677a1a51f969e25765154487bf3e85c7819db105db2fcacba83f
+size 1658691

checkpoint-1000/tokenizer_config.json

@@ -0,0 +1,16 @@
+{
+  "auto_map": {
+    "AutoTokenizer": [
+      "tokenization_internlm.InternLMTokenizer",
+      null
+    ]
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "</s>",
+  "padding_side": "right",
+  "tokenizer_class": "InternLMTokenizer",
+  "unk_token": "<unk>"
+}

checkpoint-1000/trainer_state.json

@@ -0,0 +1,616 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.9272137227630969,
+  "global_step": 1000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.999734597774032e-05,
+      "loss": 1.7592,
+      "step": 10
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.998938447446803e-05,
+      "loss": 1.6546,
+      "step": 20
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 4.997611718058365e-05,
+      "loss": 1.5632,
+      "step": 30
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.9957546913022665e-05,
+      "loss": 1.5245,
+      "step": 40
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 4.993367761465736e-05,
+      "loss": 1.5049,
+      "step": 50
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.9904514353459654e-05,
+      "loss": 1.492,
+      "step": 60
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.9870063321425105e-05,
+      "loss": 1.4803,
+      "step": 70
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 4.983033183325818e-05,
+      "loss": 1.4827,
+      "step": 80
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 4.97853283248192e-05,
+      "loss": 1.4866,
+      "step": 90
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 4.973506235133323e-05,
+      "loss": 1.486,
+      "step": 100
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 4.967954458536126e-05,
+      "loss": 1.4822,
+      "step": 110
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 4.9618786814534226e-05,
+      "loss": 1.4801,
+      "step": 120
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 4.955280193905022e-05,
+      "loss": 1.4764,
+      "step": 130
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.948160396893553e-05,
+      "loss": 1.4662,
+      "step": 140
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 4.9405208021069946e-05,
+      "loss": 1.4627,
+      "step": 150
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 4.9323630315977156e-05,
+      "loss": 1.4543,
+      "step": 160
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.9236888174380784e-05,
+      "loss": 1.4508,
+      "step": 170
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 4.91450000135268e-05,
+      "loss": 1.464,
+      "step": 180
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.9047985343273154e-05,
+      "loss": 1.4705,
+      "step": 190
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.894586476194739e-05,
+      "loss": 1.4588,
+      "step": 200
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.883865995197319e-05,
+      "loss": 1.4532,
+      "step": 210
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.8726393675266716e-05,
+      "loss": 1.4629,
+      "step": 220
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 4.860908976840376e-05,
+      "loss": 1.4429,
+      "step": 230
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 4.848677313755872e-05,
+      "loss": 1.441,
+      "step": 240
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 4.835946975321647e-05,
+      "loss": 1.4496,
+      "step": 250
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 4.822720664465827e-05,
+      "loss": 1.4459,
+      "step": 260
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 4.809001189422287e-05,
+      "loss": 1.4458,
+      "step": 270
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 4.794791463134399e-05,
+      "loss": 1.4484,
+      "step": 280
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 4.780094502636552e-05,
+      "loss": 1.4541,
+      "step": 290
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 4.764913428413572e-05,
+      "loss": 1.437,
+      "step": 300
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 4.7492514637381727e-05,
+      "loss": 1.4368,
+      "step": 310
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 4.733111933986583e-05,
+      "loss": 1.4449,
+      "step": 320
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 4.716498265932501e-05,
+      "loss": 1.4256,
+      "step": 330
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 4.699413987019512e-05,
+      "loss": 1.4378,
+      "step": 340
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 4.681862724612141e-05,
+      "loss": 1.454,
+      "step": 350
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 4.663848205225674e-05,
+      "loss": 1.4207,
+      "step": 360
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 4.645374253734949e-05,
+      "loss": 1.4286,
+      "step": 370
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 4.626444792562244e-05,
+      "loss": 1.4244,
+      "step": 380
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 4.607063840844463e-05,
+      "loss": 1.4342,
+      "step": 390
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 4.587235513579791e-05,
+      "loss": 1.4241,
+      "step": 400
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 4.5669640207539786e-05,
+      "loss": 1.443,
+      "step": 410
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 4.546253666446484e-05,
+      "loss": 1.4281,
+      "step": 420
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 4.525108847916614e-05,
+      "loss": 1.4193,
+      "step": 430
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 4.503534054669892e-05,
+      "loss": 1.4154,
+      "step": 440
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 4.481533867504841e-05,
+      "loss": 1.4273,
+      "step": 450
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 4.4591129575403765e-05,
+      "loss": 1.4329,
+      "step": 460
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 4.43627608522403e-05,
+      "loss": 1.4102,
+      "step": 470
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 4.4130280993211974e-05,
+      "loss": 1.4244,
+      "step": 480
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 4.389373935885646e-05,
+      "loss": 1.4407,
+      "step": 490
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 4.365318617211479e-05,
+      "loss": 1.4282,
+      "step": 500
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 4.340867250766794e-05,
+      "loss": 1.4269,
+      "step": 510
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 4.316025028109258e-05,
+      "loss": 1.43,
+      "step": 520
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 4.2907972237838225e-05,
+      "loss": 1.4271,
+      "step": 530
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 4.2651891942028274e-05,
+      "loss": 1.4321,
+      "step": 540
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 4.239206376508717e-05,
+      "loss": 1.4083,
+      "step": 550
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 4.212854287419611e-05,
+      "loss": 1.4168,
+      "step": 560
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 4.1861385220579934e-05,
+      "loss": 1.4144,
+      "step": 570
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 4.1590647527627404e-05,
+      "loss": 1.4106,
+      "step": 580
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 4.131638727884762e-05,
+      "loss": 1.4202,
+      "step": 590
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 4.103866270566498e-05,
+      "loss": 1.417,
+      "step": 600
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 4.075753277505544e-05,
+      "loss": 1.4324,
+      "step": 610
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 4.0473057177026484e-05,
+      "loss": 1.4147,
+      "step": 620
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 4.018529631194369e-05,
+      "loss": 1.4269,
+      "step": 630
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 3.989431127770635e-05,
+      "loss": 1.4133,
+      "step": 640
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 3.960016385677513e-05,
+      "loss": 1.4173,
+      "step": 650
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 3.9302916503054246e-05,
+      "loss": 1.4007,
+      "step": 660
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 3.9002632328631164e-05,
+      "loss": 1.4117,
+      "step": 670
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 3.8699375090376534e-05,
+      "loss": 1.4247,
+      "step": 680
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 3.8393209176407223e-05,
+      "loss": 1.4312,
+      "step": 690
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 3.8084199592415305e-05,
+      "loss": 1.4136,
+      "step": 700
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 3.777241194786591e-05,
+      "loss": 1.4132,
+      "step": 710
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 3.745791244206697e-05,
+      "loss": 1.4145,
+      "step": 720
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 3.714076785011359e-05,
+      "loss": 1.3951,
+      "step": 730
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 3.682104550871031e-05,
+      "loss": 1.4139,
+      "step": 740
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 3.649881330187401e-05,
+      "loss": 1.4078,
+      "step": 750
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 3.617413964652067e-05,
+      "loss": 1.4048,
+      "step": 760
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 3.5847093477938956e-05,
+      "loss": 1.4115,
+      "step": 770
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 3.551774423515378e-05,
+      "loss": 1.4018,
+      "step": 780
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 3.518616184618288e-05,
+      "loss": 1.4091,
+      "step": 790
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 3.4852416713189526e-05,
+      "loss": 1.4135,
+      "step": 800
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 3.4516579697534705e-05,
+      "loss": 1.3977,
+      "step": 810
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 3.417872210473162e-05,
+      "loss": 1.4123,
+      "step": 820
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 3.3838915669306034e-05,
+      "loss": 1.4096,
+      "step": 830
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 3.349723253956542e-05,
+      "loss": 1.4031,
+      "step": 840
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 3.315374526228036e-05,
+      "loss": 1.4104,
+      "step": 850
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 3.2808526767281225e-05,
+      "loss": 1.3972,
+      "step": 860
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 3.246165035197364e-05,
+      "loss": 1.3987,
+      "step": 870
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 3.211318966577581e-05,
+      "loss": 1.4197,
+      "step": 880
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 3.176321869448116e-05,
+      "loss": 1.4169,
+      "step": 890
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 3.1411811744549536e-05,
+      "loss": 1.4055,
+      "step": 900
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 3.105904342733032e-05,
+      "loss": 1.402,
+      "step": 910
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 3.070498864322081e-05,
+      "loss": 1.4148,
+      "step": 920
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 3.034972256576328e-05,
+      "loss": 1.4007,
+      "step": 930
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 2.999332062568395e-05,
+      "loss": 1.3926,
+      "step": 940
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 2.9635858494877384e-05,
+      "loss": 1.4099,
+      "step": 950
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 2.9277412070339782e-05,
+      "loss": 1.401,
+      "step": 960
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 2.891805745805429e-05,
+      "loss": 1.3995,
+      "step": 970
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 2.8557870956832132e-05,
+      "loss": 1.4108,
+      "step": 980
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 2.8196929042112652e-05,
+      "loss": 1.4006,
+      "step": 990
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 2.783530834972594e-05,
+      "loss": 1.4045,
+      "step": 1000
+    }
+  ],
+  "max_steps": 2156,
+  "num_train_epochs": 2,
+  "total_flos": 2.1503929922743173e+18,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-1000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:918a94b2b648985fdb03a840052eff79b0035343fd87ee2a0dc751e904fad0c4
+size 4091

checkpoint-2000/README.md

@@ -0,0 +1,20 @@
+---
+library_name: peft
+---
+## Training procedure
+
+
+The following `bitsandbytes` quantization config was used during training:
+- load_in_8bit: False
+- load_in_4bit: True
+- llm_int8_threshold: 6.0
+- llm_int8_skip_modules: None
+- llm_int8_enable_fp32_cpu_offload: False
+- llm_int8_has_fp16_weight: False
+- bnb_4bit_quant_type: nf4
+- bnb_4bit_use_double_quant: True
+- bnb_4bit_compute_dtype: float16
+### Framework versions
+
+
+- PEFT 0.4.0

checkpoint-2000/adapter_config.json

@@ -0,0 +1,21 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": "../internlm-chat-20b",
+  "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": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM"
+}

checkpoint-2000/adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:62fc465988009c55add9449eaec20a52187ade9f6c05f11dd25215d9acc02f03
+size 39407885

checkpoint-2000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:11a5b792ce3d6044eacf0edb6fe8dc6a5669b69aac9d1addf905381b5a71b29d
+size 78784709

checkpoint-2000/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:10784e74418dde082729bc4ccae91bab345cfe23991999f469fcec392df28c9c
+size 18679

checkpoint-2000/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:876ffe11f58c3eb52f532fa7b0a7081fa6782d9e569651792ca2ae30d57585f1
+size 18679

checkpoint-2000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dd0b604a79d16282fd4dcd9803f6bbe0802175560169f2fdb3fe364043fcb56a
+size 627

checkpoint-2000/special_tokens_map.json

@@ -0,0 +1,9 @@
+{
+  "additional_special_tokens": [
+    "<eoa>"
+  ],
+  "bos_token": "<s>",
+  "eos_token": "</s>",
+  "pad_token": "</s>",
+  "unk_token": "<unk>"
+}

checkpoint-2000/tokenization_internlm.py

@@ -0,0 +1,242 @@
+# coding=utf-8
+# Copyright 2022 EleutherAI and the HuggingFace Inc. team. All rights reserved.
+#
+# This code is based on EleutherAI's GPT-NeoX library and the GPT-NeoX
+# and OPT implementations in this library. It has been modified from its
+# original forms to accommodate minor architectural differences compared
+# to GPT-NeoX and OPT used by the Meta AI team that trained the model.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+"""Tokenization classes for IntermLM."""
+import os
+from shutil import copyfile
+from typing import Any, Dict, List, Optional, Tuple
+
+import sentencepiece as spm
+
+from transformers.tokenization_utils import PreTrainedTokenizer
+from transformers.utils import logging
+
+
+logger = logging.get_logger(__name__)
+
+VOCAB_FILES_NAMES = {"vocab_file": "./tokenizer.model"}
+
+PRETRAINED_VOCAB_FILES_MAP = {}
+
+
+class InternLMTokenizer(PreTrainedTokenizer):
+    """
+    Construct a InternLM tokenizer. Based on byte-level Byte-Pair-Encoding.
+
+    Args:
+        vocab_file (`str`):
+            Path to the vocabulary file.
+    """
+
+    vocab_files_names = VOCAB_FILES_NAMES
+    pretrained_vocab_files_map = PRETRAINED_VOCAB_FILES_MAP
+    model_input_names = ["input_ids", "attention_mask"]
+    _auto_class = "AutoTokenizer"
+
+    def __init__(
+        self,
+        vocab_file,
+        unk_token="<unk>",
+        bos_token="<s>",
+        eos_token="</s>",
+        pad_token="</s>",
+        sp_model_kwargs: Optional[Dict[str, Any]] = None,
+        add_bos_token=True,
+        add_eos_token=False,
+        decode_with_prefix_space=False,
+        clean_up_tokenization_spaces=False,
+        **kwargs,
+    ):
+        self.sp_model_kwargs = {} if sp_model_kwargs is None else sp_model_kwargs
+        super().__init__(
+            bos_token=bos_token,
+            eos_token=eos_token,
+            unk_token=unk_token,
+            pad_token=pad_token,
+            clean_up_tokenization_spaces=clean_up_tokenization_spaces,
+            **kwargs,
+        )
+        self.vocab_file = vocab_file
+        self.add_bos_token = add_bos_token
+        self.add_eos_token = add_eos_token
+        self.decode_with_prefix_space = decode_with_prefix_space
+        self.sp_model = spm.SentencePieceProcessor(**self.sp_model_kwargs)
+        self.sp_model.Load(vocab_file)
+        self._no_prefix_space_tokens = None
+
+        """ Initialisation"""
+
+    @property
+    def no_prefix_space_tokens(self):
+        if self._no_prefix_space_tokens is None:
+            vocab = self.convert_ids_to_tokens(list(range(self.vocab_size)))
+            self._no_prefix_space_tokens = {i for i, tok in enumerate(vocab) if not tok.startswith("▁")}
+        return self._no_prefix_space_tokens
+
+    @property
+    def vocab_size(self):
+        """Returns vocab size"""
+        return self.sp_model.get_piece_size()
+
+    @property
+    def bos_token_id(self) -> Optional[int]:
+        return self.sp_model.bos_id()
+
+    @property
+    def eos_token_id(self) -> Optional[int]:
+        return self.sp_model.eos_id()
+
+    def get_vocab(self):
+        """Returns vocab as a dict"""
+        vocab = {self.convert_ids_to_tokens(i): i for i in range(self.vocab_size)}
+        vocab.update(self.added_tokens_encoder)
+        return vocab
+
+    def _tokenize(self, text):
+        """Returns a tokenized string."""
+        return self.sp_model.encode(text, out_type=str)
+
+    def _convert_token_to_id(self, token):
+        """Converts a token (str) in an id using the vocab."""
+        return self.sp_model.piece_to_id(token)
+
+    def _convert_id_to_token(self, index):
+        """Converts an index (integer) in a token (str) using the vocab."""
+        token = self.sp_model.IdToPiece(index)
+        return token
+
+    def _maybe_add_prefix_space(self, tokens, decoded):
+        if tokens and tokens[0] not in self.no_prefix_space_tokens:
+            return " " + decoded
+        else:
+            return decoded
+
+    def convert_tokens_to_string(self, tokens):
+        """Converts a sequence of tokens (string) in a single string."""
+        current_sub_tokens = []
+        out_string = ""
+        prev_is_special = False
+        for token in tokens:
+            # make sure that special tokens are not decoded using sentencepiece model
+            if token in self.all_special_tokens:
+                if not prev_is_special:
+                    out_string += " "
+                out_string += self.sp_model.decode(current_sub_tokens) + token
+                prev_is_special = True
+                current_sub_tokens = []
+            else:
+                current_sub_tokens.append(token)
+                prev_is_special = False
+        out_string += self.sp_model.decode(current_sub_tokens)
+        out_string = self.clean_up_tokenization(out_string)
+        out_string = self._maybe_add_prefix_space(tokens=tokens, decoded=out_string)
+        return out_string[1:]
+
+    def save_vocabulary(self, save_directory, filename_prefix: Optional[str] = None) -> Tuple[str]:
+        """
+        Save the vocabulary and special tokens file to a directory.
+
+        Args:
+            save_directory (`str`):
+                The directory in which to save the vocabulary.
+
+        Returns:
+            `Tuple(str)`: Paths to the files saved.
+        """
+        if not os.path.isdir(save_directory):
+            logger.error(f"Vocabulary path ({save_directory}) should be a directory")
+            return
+        out_vocab_file = os.path.join(
+            save_directory, (filename_prefix + "-" if filename_prefix else "") + VOCAB_FILES_NAMES["vocab_file"]
+        )
+
+        if os.path.abspath(self.vocab_file) != os.path.abspath(out_vocab_file) and os.path.isfile(self.vocab_file):
+            copyfile(self.vocab_file, out_vocab_file)
+        elif not os.path.isfile(self.vocab_file):
+            with open(out_vocab_file, "wb") as fi:
+                content_spiece_model = self.sp_model.serialized_model_proto()
+                fi.write(content_spiece_model)
+
+        return (out_vocab_file,)
+
+    def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=None):
+        if self.add_bos_token:
+            bos_token_ids = [self.bos_token_id]
+        else:
+            bos_token_ids = []
+
+        output = bos_token_ids + token_ids_0
+
+        if token_ids_1 is not None:
+            output = output + token_ids_1
+
+        if self.add_eos_token:
+            output = output + [self.eos_token_id]
+
+        return output
+
+    def get_special_tokens_mask(
+        self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None, already_has_special_tokens: bool = False
+    ) -> List[int]:
+        """
+        Retrieve sequence ids from a token list that has no special tokens added. This method is called when adding
+        special tokens using the tokenizer `prepare_for_model` method.
+
+        Args:
+            token_ids_0 (`List[int]`):
+                List of IDs.
+            token_ids_1 (`List[int]`, *optional*):
+                Optional second list of IDs for sequence pairs.
+            already_has_special_tokens (`bool`, *optional*, defaults to `False`):
+                Whether or not the token list is already formatted with special tokens for the model.
+
+        Returns:
+            `List[int]`: A list of integers in the range [0, 1]: 1 for a special token, 0 for a sequence token.
+        """
+        if already_has_special_tokens:
+            return super().get_special_tokens_mask(
+                token_ids_0=token_ids_0, token_ids_1=token_ids_1, already_has_special_tokens=True
+            )
+
+        if token_ids_1 is None:
+            return [1] + ([0] * len(token_ids_0)) + [1]
+        return [1] + ([0] * len(token_ids_0)) + [1, 1] + ([0] * len(token_ids_1)) + [1]
+
+    def create_token_type_ids_from_sequences(
+        self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None
+    ) -> List[int]:
+        """
+        Create a mask from the two sequences passed to be used in a sequence-pair classification task. T5 does not make
+        use of token type ids, therefore a list of zeros is returned.
+
+        Args:
+            token_ids_0 (`List[int]`):
+                List of IDs.
+            token_ids_1 (`List[int]`, *optional*):
+                Optional second list of IDs for sequence pairs.
+
+        Returns:
+            `List[int]`: List of zeros.
+        """
+        eos = [self.eos_token_id]
+
+        if token_ids_1 is None:
+            return len(token_ids_0 + eos) * [0]
+        return len(token_ids_0 + eos + token_ids_1 + eos) * [0]

checkpoint-2000/tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aab622d98c98677a1a51f969e25765154487bf3e85c7819db105db2fcacba83f
+size 1658691

checkpoint-2000/tokenizer_config.json

@@ -0,0 +1,16 @@
+{
+  "auto_map": {
+    "AutoTokenizer": [
+      "tokenization_internlm.InternLMTokenizer",
+      null
+    ]
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "</s>",
+  "padding_side": "right",
+  "tokenizer_class": "InternLMTokenizer",
+  "unk_token": "<unk>"
+}

checkpoint-2000/trainer_state.json

@@ -0,0 +1,1216 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.8544274455261938,
+  "global_step": 2000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.999734597774032e-05,
+      "loss": 1.7592,
+      "step": 10
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.998938447446803e-05,
+      "loss": 1.6546,
+      "step": 20
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 4.997611718058365e-05,
+      "loss": 1.5632,
+      "step": 30
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.9957546913022665e-05,
+      "loss": 1.5245,
+      "step": 40
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 4.993367761465736e-05,
+      "loss": 1.5049,
+      "step": 50
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.9904514353459654e-05,
+      "loss": 1.492,
+      "step": 60
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.9870063321425105e-05,
+      "loss": 1.4803,
+      "step": 70
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 4.983033183325818e-05,
+      "loss": 1.4827,
+      "step": 80
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 4.97853283248192e-05,
+      "loss": 1.4866,
+      "step": 90
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 4.973506235133323e-05,
+      "loss": 1.486,
+      "step": 100
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 4.967954458536126e-05,
+      "loss": 1.4822,
+      "step": 110
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 4.9618786814534226e-05,
+      "loss": 1.4801,
+      "step": 120
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 4.955280193905022e-05,
+      "loss": 1.4764,
+      "step": 130
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.948160396893553e-05,
+      "loss": 1.4662,
+      "step": 140
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 4.9405208021069946e-05,
+      "loss": 1.4627,
+      "step": 150
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 4.9323630315977156e-05,
+      "loss": 1.4543,
+      "step": 160
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.9236888174380784e-05,
+      "loss": 1.4508,
+      "step": 170
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 4.91450000135268e-05,
+      "loss": 1.464,
+      "step": 180
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.9047985343273154e-05,
+      "loss": 1.4705,
+      "step": 190
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.894586476194739e-05,
+      "loss": 1.4588,
+      "step": 200
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.883865995197319e-05,
+      "loss": 1.4532,
+      "step": 210
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.8726393675266716e-05,
+      "loss": 1.4629,
+      "step": 220
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 4.860908976840376e-05,
+      "loss": 1.4429,
+      "step": 230
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 4.848677313755872e-05,
+      "loss": 1.441,
+      "step": 240
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 4.835946975321647e-05,
+      "loss": 1.4496,
+      "step": 250
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 4.822720664465827e-05,
+      "loss": 1.4459,
+      "step": 260
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 4.809001189422287e-05,
+      "loss": 1.4458,
+      "step": 270
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 4.794791463134399e-05,
+      "loss": 1.4484,
+      "step": 280
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 4.780094502636552e-05,
+      "loss": 1.4541,
+      "step": 290
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 4.764913428413572e-05,
+      "loss": 1.437,
+      "step": 300
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 4.7492514637381727e-05,
+      "loss": 1.4368,
+      "step": 310
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 4.733111933986583e-05,
+      "loss": 1.4449,
+      "step": 320
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 4.716498265932501e-05,
+      "loss": 1.4256,
+      "step": 330
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 4.699413987019512e-05,
+      "loss": 1.4378,
+      "step": 340
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 4.681862724612141e-05,
+      "loss": 1.454,
+      "step": 350
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 4.663848205225674e-05,
+      "loss": 1.4207,
+      "step": 360
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 4.645374253734949e-05,
+      "loss": 1.4286,
+      "step": 370
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 4.626444792562244e-05,
+      "loss": 1.4244,
+      "step": 380
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 4.607063840844463e-05,
+      "loss": 1.4342,
+      "step": 390
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 4.587235513579791e-05,
+      "loss": 1.4241,
+      "step": 400
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 4.5669640207539786e-05,
+      "loss": 1.443,
+      "step": 410
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 4.546253666446484e-05,
+      "loss": 1.4281,
+      "step": 420
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 4.525108847916614e-05,
+      "loss": 1.4193,
+      "step": 430
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 4.503534054669892e-05,
+      "loss": 1.4154,
+      "step": 440
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 4.481533867504841e-05,
+      "loss": 1.4273,
+      "step": 450
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 4.4591129575403765e-05,
+      "loss": 1.4329,
+      "step": 460
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 4.43627608522403e-05,
+      "loss": 1.4102,
+      "step": 470
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 4.4130280993211974e-05,
+      "loss": 1.4244,
+      "step": 480
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 4.389373935885646e-05,
+      "loss": 1.4407,
+      "step": 490
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 4.365318617211479e-05,
+      "loss": 1.4282,
+      "step": 500
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 4.340867250766794e-05,
+      "loss": 1.4269,
+      "step": 510
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 4.316025028109258e-05,
+      "loss": 1.43,
+      "step": 520
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 4.2907972237838225e-05,
+      "loss": 1.4271,
+      "step": 530
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 4.2651891942028274e-05,
+      "loss": 1.4321,
+      "step": 540
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 4.239206376508717e-05,
+      "loss": 1.4083,
+      "step": 550
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 4.212854287419611e-05,
+      "loss": 1.4168,
+      "step": 560
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 4.1861385220579934e-05,
+      "loss": 1.4144,
+      "step": 570
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 4.1590647527627404e-05,
+      "loss": 1.4106,
+      "step": 580
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 4.131638727884762e-05,
+      "loss": 1.4202,
+      "step": 590
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 4.103866270566498e-05,
+      "loss": 1.417,
+      "step": 600
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 4.075753277505544e-05,
+      "loss": 1.4324,
+      "step": 610
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 4.0473057177026484e-05,
+      "loss": 1.4147,
+      "step": 620
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 4.018529631194369e-05,
+      "loss": 1.4269,
+      "step": 630
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 3.989431127770635e-05,
+      "loss": 1.4133,
+      "step": 640
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 3.960016385677513e-05,
+      "loss": 1.4173,
+      "step": 650
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 3.9302916503054246e-05,
+      "loss": 1.4007,
+      "step": 660
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 3.9002632328631164e-05,
+      "loss": 1.4117,
+      "step": 670
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 3.8699375090376534e-05,
+      "loss": 1.4247,
+      "step": 680
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 3.8393209176407223e-05,
+      "loss": 1.4312,
+      "step": 690
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 3.8084199592415305e-05,
+      "loss": 1.4136,
+      "step": 700
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 3.777241194786591e-05,
+      "loss": 1.4132,
+      "step": 710
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 3.745791244206697e-05,
+      "loss": 1.4145,
+      "step": 720
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 3.714076785011359e-05,
+      "loss": 1.3951,
+      "step": 730
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 3.682104550871031e-05,
+      "loss": 1.4139,
+      "step": 740
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 3.649881330187401e-05,
+      "loss": 1.4078,
+      "step": 750
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 3.617413964652067e-05,
+      "loss": 1.4048,
+      "step": 760
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 3.5847093477938956e-05,
+      "loss": 1.4115,
+      "step": 770
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 3.551774423515378e-05,
+      "loss": 1.4018,
+      "step": 780
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 3.518616184618288e-05,
+      "loss": 1.4091,
+      "step": 790
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 3.4852416713189526e-05,
+      "loss": 1.4135,
+      "step": 800
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 3.4516579697534705e-05,
+      "loss": 1.3977,
+      "step": 810
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 3.417872210473162e-05,
+      "loss": 1.4123,
+      "step": 820
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 3.3838915669306034e-05,
+      "loss": 1.4096,
+      "step": 830
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 3.349723253956542e-05,
+      "loss": 1.4031,
+      "step": 840
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 3.315374526228036e-05,
+      "loss": 1.4104,
+      "step": 850
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 3.2808526767281225e-05,
+      "loss": 1.3972,
+      "step": 860
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 3.246165035197364e-05,
+      "loss": 1.3987,
+      "step": 870
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 3.211318966577581e-05,
+      "loss": 1.4197,
+      "step": 880
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 3.176321869448116e-05,
+      "loss": 1.4169,
+      "step": 890
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 3.1411811744549536e-05,
+      "loss": 1.4055,
+      "step": 900
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 3.105904342733032e-05,
+      "loss": 1.402,
+      "step": 910
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 3.070498864322081e-05,
+      "loss": 1.4148,
+      "step": 920
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 3.034972256576328e-05,
+      "loss": 1.4007,
+      "step": 930
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 2.999332062568395e-05,
+      "loss": 1.3926,
+      "step": 940
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 2.9635858494877384e-05,
+      "loss": 1.4099,
+      "step": 950
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 2.9277412070339782e-05,
+      "loss": 1.401,
+      "step": 960
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 2.891805745805429e-05,
+      "loss": 1.3995,
+      "step": 970
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 2.8557870956832132e-05,
+      "loss": 1.4108,
+      "step": 980
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 2.8196929042112652e-05,
+      "loss": 1.4006,
+      "step": 990
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 2.783530834972594e-05,
+      "loss": 1.4045,
+      "step": 1000
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 2.7473085659621377e-05,
+      "loss": 1.3907,
+      "step": 1010
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 2.711033787956555e-05,
+      "loss": 1.3872,
+      "step": 1020
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 2.6747142028813105e-05,
+      "loss": 1.405,
+      "step": 1030
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 2.638357522175383e-05,
+      "loss": 1.3952,
+      "step": 1040
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 2.6019714651539646e-05,
+      "loss": 1.3901,
+      "step": 1050
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 2.565563757369475e-05,
+      "loss": 1.4027,
+      "step": 1060
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 2.529142128971268e-05,
+      "loss": 1.3909,
+      "step": 1070
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.492714313064342e-05,
+      "loss": 1.4039,
+      "step": 1080
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.45628804406744e-05,
+      "loss": 1.3964,
+      "step": 1090
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.419871056070862e-05,
+      "loss": 1.4076,
+      "step": 1100
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 2.3834710811943514e-05,
+      "loss": 1.3891,
+      "step": 1110
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 2.3470958479453938e-05,
+      "loss": 1.4004,
+      "step": 1120
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 2.3107530795782877e-05,
+      "loss": 1.3879,
+      "step": 1130
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 2.2744504924543313e-05,
+      "loss": 1.3879,
+      "step": 1140
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 2.23819579440347e-05,
+      "loss": 1.3851,
+      "step": 1150
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 2.2019966830877545e-05,
+      "loss": 1.3846,
+      "step": 1160
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 2.1658608443669635e-05,
+      "loss": 1.3936,
+      "step": 1170
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 2.1297959506667224e-05,
+      "loss": 1.3841,
+      "step": 1180
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 2.0938096593494855e-05,
+      "loss": 1.3969,
+      "step": 1190
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 2.057909611088709e-05,
+      "loss": 1.3953,
+      "step": 1200
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 2.02210342824657e-05,
+      "loss": 1.3951,
+      "step": 1210
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.9863987132555706e-05,
+      "loss": 1.4012,
+      "step": 1220
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.9508030470043806e-05,
+      "loss": 1.3998,
+      "step": 1230
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 1.915323987228247e-05,
+      "loss": 1.3933,
+      "step": 1240
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.8799690669043212e-05,
+      "loss": 1.3977,
+      "step": 1250
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 1.8447457926522454e-05,
+      "loss": 1.3818,
+      "step": 1260
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.8096616431403325e-05,
+      "loss": 1.4032,
+      "step": 1270
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.7747240674976857e-05,
+      "loss": 1.3853,
+      "step": 1280
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.7399404837325796e-05,
+      "loss": 1.3988,
+      "step": 1290
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.7053182771574633e-05,
+      "loss": 1.3963,
+      "step": 1300
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.6708647988208887e-05,
+      "loss": 1.3881,
+      "step": 1310
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.6365873639467315e-05,
+      "loss": 1.3804,
+      "step": 1320
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.602493250381003e-05,
+      "loss": 1.3952,
+      "step": 1330
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.5685896970466123e-05,
+      "loss": 1.4015,
+      "step": 1340
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 1.534883902406375e-05,
+      "loss": 1.3797,
+      "step": 1350
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.5013830229346326e-05,
+      "loss": 1.3931,
+      "step": 1360
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.4680941715977722e-05,
+      "loss": 1.3946,
+      "step": 1370
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 1.4350244163439892e-05,
+      "loss": 1.3851,
+      "step": 1380
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.4021807786026108e-05,
+      "loss": 1.3891,
+      "step": 1390
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.3695702317932862e-05,
+      "loss": 1.3958,
+      "step": 1400
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.337199699845387e-05,
+      "loss": 1.3909,
+      "step": 1410
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.3050760557279023e-05,
+      "loss": 1.3785,
+      "step": 1420
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 1.2732061199901562e-05,
+      "loss": 1.3882,
+      "step": 1430
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.2415966593136547e-05,
+      "loss": 1.3872,
+      "step": 1440
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.2102543850753808e-05,
+      "loss": 1.4032,
+      "step": 1450
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.1791859519228138e-05,
+      "loss": 1.3928,
+      "step": 1460
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 1.148397956361007e-05,
+      "loss": 1.3785,
+      "step": 1470
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 1.1178969353520018e-05,
+      "loss": 1.3797,
+      "step": 1480
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.087689364926897e-05,
+      "loss": 1.3966,
+      "step": 1490
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.0577816588108378e-05,
+      "loss": 1.3849,
+      "step": 1500
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 1.0281801670612449e-05,
+      "loss": 1.3732,
+      "step": 1510
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 9.988911747195603e-06,
+      "loss": 1.4018,
+      "step": 1520
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 9.699209004767953e-06,
+      "loss": 1.3902,
+      "step": 1530
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 9.412754953531663e-06,
+      "loss": 1.3991,
+      "step": 1540
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 9.129610413921005e-06,
+      "loss": 1.3824,
+      "step": 1550
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 8.849835503688847e-06,
+      "loss": 1.3904,
+      "step": 1560
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 8.57348962514234e-06,
+      "loss": 1.3937,
+      "step": 1570
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 8.30063145253053e-06,
+      "loss": 1.39,
+      "step": 1580
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 8.031318919586523e-06,
+      "loss": 1.3773,
+      "step": 1590
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 7.7656092072269e-06,
+      "loss": 1.3875,
+      "step": 1600
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 7.503558731410959e-06,
+      "loss": 1.3859,
+      "step": 1610
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 7.245223131162376e-06,
+      "loss": 1.3716,
+      "step": 1620
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 6.9906572567558285e-06,
+      "loss": 1.3846,
+      "step": 1630
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 6.739915158071106e-06,
+      "loss": 1.3741,
+      "step": 1640
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 6.493050073117116e-06,
+      "loss": 1.3827,
+      "step": 1650
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 6.250114416728298e-06,
+      "loss": 1.3951,
+      "step": 1660
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 6.011159769435823e-06,
+      "loss": 1.3875,
+      "step": 1670
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 5.776236866515947e-06,
+      "loss": 1.3783,
+      "step": 1680
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 5.545395587217763e-06,
+      "loss": 1.3888,
+      "step": 1690
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 5.318684944172752e-06,
+      "loss": 1.4006,
+      "step": 1700
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 5.096153072988366e-06,
+      "loss": 1.4041,
+      "step": 1710
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 4.8778472220277874e-06,
+      "loss": 1.3865,
+      "step": 1720
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 4.6638137423780395e-06,
+      "loss": 1.3789,
+      "step": 1730
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 4.454098078008667e-06,
+      "loss": 1.3861,
+      "step": 1740
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 4.248744756122986e-06,
+      "loss": 1.3803,
+      "step": 1750
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 4.047797377703985e-06,
+      "loss": 1.3695,
+      "step": 1760
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 3.851298608256892e-06,
+      "loss": 1.3937,
+      "step": 1770
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 3.6592901687503566e-06,
+      "loss": 1.3865,
+      "step": 1780
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 3.471812826758178e-06,
+      "loss": 1.3799,
+      "step": 1790
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 3.288906387803464e-06,
+      "loss": 1.3911,
+      "step": 1800
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 3.1106096869070483e-06,
+      "loss": 1.3801,
+      "step": 1810
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 2.9369605803419715e-06,
+      "loss": 1.382,
+      "step": 1820
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 2.767995937595766e-06,
+      "loss": 1.3898,
+      "step": 1830
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 2.6037516335422728e-06,
+      "loss": 1.3752,
+      "step": 1840
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 2.4442625408246074e-06,
+      "loss": 1.3855,
+      "step": 1850
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 2.289562522450947e-06,
+      "loss": 1.3665,
+      "step": 1860
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 2.1396844246046903e-06,
+      "loss": 1.3747,
+      "step": 1870
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 1.9946600696704592e-06,
+      "loss": 1.3861,
+      "step": 1880
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 1.8545202494775509e-06,
+      "loss": 1.3844,
+      "step": 1890
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 1.7192947187621434e-06,
+      "loss": 1.3796,
+      "step": 1900
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 1.5890121888497366e-06,
+      "loss": 1.3963,
+      "step": 1910
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 1.463700321559075e-06,
+      "loss": 1.385,
+      "step": 1920
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 1.3433857233289714e-06,
+      "loss": 1.3811,
+      "step": 1930
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 1.2280939395691859e-06,
+      "loss": 1.384,
+      "step": 1940
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 1.1178494492365465e-06,
+      "loss": 1.3821,
+      "step": 1950
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 1.0126756596375686e-06,
+      "loss": 1.3978,
+      "step": 1960
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 9.125949014585383e-07,
+      "loss": 1.3724,
+      "step": 1970
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 8.176284240242638e-07,
+      "loss": 1.3851,
+      "step": 1980
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 7.277963907863478e-07,
+      "loss": 1.3793,
+      "step": 1990
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 6.431178750420513e-07,
+      "loss": 1.3946,
+      "step": 2000
+    }
+  ],
+  "max_steps": 2156,
+  "num_train_epochs": 2,
+  "total_flos": 4.3007034481621074e+18,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-2000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:918a94b2b648985fdb03a840052eff79b0035343fd87ee2a0dc751e904fad0c4
+size 4091

special_tokens_map.json

@@ -0,0 +1,9 @@
+{
+  "additional_special_tokens": [
+    "<eoa>"
+  ],
+  "bos_token": "<s>",
+  "eos_token": "</s>",
+  "pad_token": "</s>",
+  "unk_token": "<unk>"
+}

tokenization_internlm.py

@@ -0,0 +1,242 @@
+# coding=utf-8
+# Copyright 2022 EleutherAI and the HuggingFace Inc. team. All rights reserved.
+#
+# This code is based on EleutherAI's GPT-NeoX library and the GPT-NeoX
+# and OPT implementations in this library. It has been modified from its
+# original forms to accommodate minor architectural differences compared
+# to GPT-NeoX and OPT used by the Meta AI team that trained the model.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+"""Tokenization classes for IntermLM."""
+import os
+from shutil import copyfile
+from typing import Any, Dict, List, Optional, Tuple
+
+import sentencepiece as spm
+
+from transformers.tokenization_utils import PreTrainedTokenizer
+from transformers.utils import logging
+
+
+logger = logging.get_logger(__name__)
+
+VOCAB_FILES_NAMES = {"vocab_file": "./tokenizer.model"}
+
+PRETRAINED_VOCAB_FILES_MAP = {}
+
+
+class InternLMTokenizer(PreTrainedTokenizer):
+    """
+    Construct a InternLM tokenizer. Based on byte-level Byte-Pair-Encoding.
+
+    Args:
+        vocab_file (`str`):
+            Path to the vocabulary file.
+    """
+
+    vocab_files_names = VOCAB_FILES_NAMES
+    pretrained_vocab_files_map = PRETRAINED_VOCAB_FILES_MAP
+    model_input_names = ["input_ids", "attention_mask"]
+    _auto_class = "AutoTokenizer"
+
+    def __init__(
+        self,
+        vocab_file,
+        unk_token="<unk>",
+        bos_token="<s>",
+        eos_token="</s>",
+        pad_token="</s>",
+        sp_model_kwargs: Optional[Dict[str, Any]] = None,
+        add_bos_token=True,
+        add_eos_token=False,
+        decode_with_prefix_space=False,
+        clean_up_tokenization_spaces=False,
+        **kwargs,
+    ):
+        self.sp_model_kwargs = {} if sp_model_kwargs is None else sp_model_kwargs
+        super().__init__(
+            bos_token=bos_token,
+            eos_token=eos_token,
+            unk_token=unk_token,
+            pad_token=pad_token,
+            clean_up_tokenization_spaces=clean_up_tokenization_spaces,
+            **kwargs,
+        )
+        self.vocab_file = vocab_file
+        self.add_bos_token = add_bos_token
+        self.add_eos_token = add_eos_token
+        self.decode_with_prefix_space = decode_with_prefix_space
+        self.sp_model = spm.SentencePieceProcessor(**self.sp_model_kwargs)
+        self.sp_model.Load(vocab_file)
+        self._no_prefix_space_tokens = None
+
+        """ Initialisation"""
+
+    @property
+    def no_prefix_space_tokens(self):
+        if self._no_prefix_space_tokens is None:
+            vocab = self.convert_ids_to_tokens(list(range(self.vocab_size)))
+            self._no_prefix_space_tokens = {i for i, tok in enumerate(vocab) if not tok.startswith("▁")}
+        return self._no_prefix_space_tokens
+
+    @property
+    def vocab_size(self):
+        """Returns vocab size"""
+        return self.sp_model.get_piece_size()
+
+    @property
+    def bos_token_id(self) -> Optional[int]:
+        return self.sp_model.bos_id()
+
+    @property
+    def eos_token_id(self) -> Optional[int]:
+        return self.sp_model.eos_id()
+
+    def get_vocab(self):
+        """Returns vocab as a dict"""
+        vocab = {self.convert_ids_to_tokens(i): i for i in range(self.vocab_size)}
+        vocab.update(self.added_tokens_encoder)
+        return vocab
+
+    def _tokenize(self, text):
+        """Returns a tokenized string."""
+        return self.sp_model.encode(text, out_type=str)
+
+    def _convert_token_to_id(self, token):
+        """Converts a token (str) in an id using the vocab."""
+        return self.sp_model.piece_to_id(token)
+
+    def _convert_id_to_token(self, index):
+        """Converts an index (integer) in a token (str) using the vocab."""
+        token = self.sp_model.IdToPiece(index)
+        return token
+
+    def _maybe_add_prefix_space(self, tokens, decoded):
+        if tokens and tokens[0] not in self.no_prefix_space_tokens:
+            return " " + decoded
+        else:
+            return decoded
+
+    def convert_tokens_to_string(self, tokens):
+        """Converts a sequence of tokens (string) in a single string."""
+        current_sub_tokens = []
+        out_string = ""
+        prev_is_special = False
+        for token in tokens:
+            # make sure that special tokens are not decoded using sentencepiece model
+            if token in self.all_special_tokens:
+                if not prev_is_special:
+                    out_string += " "
+                out_string += self.sp_model.decode(current_sub_tokens) + token
+                prev_is_special = True
+                current_sub_tokens = []
+            else:
+                current_sub_tokens.append(token)
+                prev_is_special = False
+        out_string += self.sp_model.decode(current_sub_tokens)
+        out_string = self.clean_up_tokenization(out_string)
+        out_string = self._maybe_add_prefix_space(tokens=tokens, decoded=out_string)
+        return out_string[1:]
+
+    def save_vocabulary(self, save_directory, filename_prefix: Optional[str] = None) -> Tuple[str]:
+        """
+        Save the vocabulary and special tokens file to a directory.
+
+        Args:
+            save_directory (`str`):
+                The directory in which to save the vocabulary.
+
+        Returns:
+            `Tuple(str)`: Paths to the files saved.
+        """
+        if not os.path.isdir(save_directory):
+            logger.error(f"Vocabulary path ({save_directory}) should be a directory")
+            return
+        out_vocab_file = os.path.join(
+            save_directory, (filename_prefix + "-" if filename_prefix else "") + VOCAB_FILES_NAMES["vocab_file"]
+        )
+
+        if os.path.abspath(self.vocab_file) != os.path.abspath(out_vocab_file) and os.path.isfile(self.vocab_file):
+            copyfile(self.vocab_file, out_vocab_file)
+        elif not os.path.isfile(self.vocab_file):
+            with open(out_vocab_file, "wb") as fi:
+                content_spiece_model = self.sp_model.serialized_model_proto()
+                fi.write(content_spiece_model)
+
+        return (out_vocab_file,)
+
+    def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=None):
+        if self.add_bos_token:
+            bos_token_ids = [self.bos_token_id]
+        else:
+            bos_token_ids = []
+
+        output = bos_token_ids + token_ids_0
+
+        if token_ids_1 is not None:
+            output = output + token_ids_1
+
+        if self.add_eos_token:
+            output = output + [self.eos_token_id]
+
+        return output
+
+    def get_special_tokens_mask(
+        self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None, already_has_special_tokens: bool = False
+    ) -> List[int]:
+        """
+        Retrieve sequence ids from a token list that has no special tokens added. This method is called when adding
+        special tokens using the tokenizer `prepare_for_model` method.
+
+        Args:
+            token_ids_0 (`List[int]`):
+                List of IDs.
+            token_ids_1 (`List[int]`, *optional*):
+                Optional second list of IDs for sequence pairs.
+            already_has_special_tokens (`bool`, *optional*, defaults to `False`):
+                Whether or not the token list is already formatted with special tokens for the model.
+
+        Returns:
+            `List[int]`: A list of integers in the range [0, 1]: 1 for a special token, 0 for a sequence token.
+        """
+        if already_has_special_tokens:
+            return super().get_special_tokens_mask(
+                token_ids_0=token_ids_0, token_ids_1=token_ids_1, already_has_special_tokens=True
+            )
+
+        if token_ids_1 is None:
+            return [1] + ([0] * len(token_ids_0)) + [1]
+        return [1] + ([0] * len(token_ids_0)) + [1, 1] + ([0] * len(token_ids_1)) + [1]
+
+    def create_token_type_ids_from_sequences(
+        self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None
+    ) -> List[int]:
+        """
+        Create a mask from the two sequences passed to be used in a sequence-pair classification task. T5 does not make
+        use of token type ids, therefore a list of zeros is returned.
+
+        Args:
+            token_ids_0 (`List[int]`):
+                List of IDs.
+            token_ids_1 (`List[int]`, *optional*):
+                Optional second list of IDs for sequence pairs.
+
+        Returns:
+            `List[int]`: List of zeros.
+        """
+        eos = [self.eos_token_id]
+
+        if token_ids_1 is None:
+            return len(token_ids_0 + eos) * [0]
+        return len(token_ids_0 + eos + token_ids_1 + eos) * [0]

tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aab622d98c98677a1a51f969e25765154487bf3e85c7819db105db2fcacba83f
+size 1658691

tokenizer_config.json

@@ -0,0 +1,16 @@
+{
+  "auto_map": {
+    "AutoTokenizer": [
+      "tokenization_internlm.InternLMTokenizer",
+      null
+    ]
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "</s>",
+  "padding_side": "right",
+  "tokenizer_class": "InternLMTokenizer",
+  "unk_token": "<unk>"
+}

train_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 2.0,
+    "train_loss": 1.4111853902105498,
+    "train_runtime": 51425.1791,
+    "train_samples_per_second": 2.684,
+    "train_steps_per_second": 0.042
+}

trainer_log.jsonl

@@ -0,0 +1,216 @@
+{"current_steps": 10, "total_steps": 2156, "loss": 1.7592, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.999734597774032e-05, "epoch": 0.01, "percentage": 0.46, "elapsed_time": "0:03:56", "remaining_time": "14:04:54"}
+{"current_steps": 20, "total_steps": 2156, "loss": 1.6546, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.998938447446803e-05, "epoch": 0.02, "percentage": 0.93, "elapsed_time": "0:07:53", "remaining_time": "14:03:23"}
+{"current_steps": 30, "total_steps": 2156, "loss": 1.5632, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.997611718058365e-05, "epoch": 0.03, "percentage": 1.39, "elapsed_time": "0:11:45", "remaining_time": "13:53:05"}
+{"current_steps": 40, "total_steps": 2156, "loss": 1.5245, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.9957546913022665e-05, "epoch": 0.04, "percentage": 1.86, "elapsed_time": "0:15:44", "remaining_time": "13:52:29"}
+{"current_steps": 50, "total_steps": 2156, "loss": 1.5049, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.993367761465736e-05, "epoch": 0.05, "percentage": 2.32, "elapsed_time": "0:19:48", "remaining_time": "13:54:22"}
+{"current_steps": 60, "total_steps": 2156, "loss": 1.492, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.9904514353459654e-05, "epoch": 0.06, "percentage": 2.78, "elapsed_time": "0:23:41", "remaining_time": "13:47:41"}
+{"current_steps": 70, "total_steps": 2156, "loss": 1.4803, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.9870063321425105e-05, "epoch": 0.06, "percentage": 3.25, "elapsed_time": "0:27:43", "remaining_time": "13:46:02"}
+{"current_steps": 80, "total_steps": 2156, "loss": 1.4827, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.983033183325818e-05, "epoch": 0.07, "percentage": 3.71, "elapsed_time": "0:31:38", "remaining_time": "13:40:57"}
+{"current_steps": 90, "total_steps": 2156, "loss": 1.4866, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.97853283248192e-05, "epoch": 0.08, "percentage": 4.17, "elapsed_time": "0:35:30", "remaining_time": "13:35:03"}
+{"current_steps": 100, "total_steps": 2156, "loss": 1.486, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.973506235133323e-05, "epoch": 0.09, "percentage": 4.64, "elapsed_time": "0:39:25", "remaining_time": "13:30:27"}
+{"current_steps": 110, "total_steps": 2156, "loss": 1.4822, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.967954458536126e-05, "epoch": 0.1, "percentage": 5.1, "elapsed_time": "0:43:22", "remaining_time": "13:26:52"}
+{"current_steps": 120, "total_steps": 2156, "loss": 1.4801, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.9618786814534226e-05, "epoch": 0.11, "percentage": 5.57, "elapsed_time": "0:47:19", "remaining_time": "13:22:52"}
+{"current_steps": 130, "total_steps": 2156, "loss": 1.4764, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.955280193905022e-05, "epoch": 0.12, "percentage": 6.03, "elapsed_time": "0:51:16", "remaining_time": "13:19:07"}
+{"current_steps": 140, "total_steps": 2156, "loss": 1.4662, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.948160396893553e-05, "epoch": 0.13, "percentage": 6.49, "elapsed_time": "0:55:09", "remaining_time": "13:14:11"}
+{"current_steps": 150, "total_steps": 2156, "loss": 1.4627, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.9405208021069946e-05, "epoch": 0.14, "percentage": 6.96, "elapsed_time": "0:59:02", "remaining_time": "13:09:34"}
+{"current_steps": 160, "total_steps": 2156, "loss": 1.4543, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.9323630315977156e-05, "epoch": 0.15, "percentage": 7.42, "elapsed_time": "1:03:04", "remaining_time": "13:06:48"}
+{"current_steps": 170, "total_steps": 2156, "loss": 1.4508, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.9236888174380784e-05, "epoch": 0.16, "percentage": 7.88, "elapsed_time": "1:06:57", "remaining_time": "13:02:15"}
+{"current_steps": 180, "total_steps": 2156, "loss": 1.464, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.91450000135268e-05, "epoch": 0.17, "percentage": 8.35, "elapsed_time": "1:10:51", "remaining_time": "12:57:47"}
+{"current_steps": 190, "total_steps": 2156, "loss": 1.4705, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.9047985343273154e-05, "epoch": 0.18, "percentage": 8.81, "elapsed_time": "1:14:48", "remaining_time": "12:54:02"}
+{"current_steps": 200, "total_steps": 2156, "loss": 1.4588, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.894586476194739e-05, "epoch": 0.19, "percentage": 9.28, "elapsed_time": "1:18:40", "remaining_time": "12:49:30"}
+{"current_steps": 210, "total_steps": 2156, "loss": 1.4532, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.883865995197319e-05, "epoch": 0.19, "percentage": 9.74, "elapsed_time": "1:22:38", "remaining_time": "12:45:48"}
+{"current_steps": 220, "total_steps": 2156, "loss": 1.4629, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.8726393675266716e-05, "epoch": 0.2, "percentage": 10.2, "elapsed_time": "1:26:32", "remaining_time": "12:41:37"}
+{"current_steps": 230, "total_steps": 2156, "loss": 1.4429, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.860908976840376e-05, "epoch": 0.21, "percentage": 10.67, "elapsed_time": "1:30:23", "remaining_time": "12:36:52"}
+{"current_steps": 240, "total_steps": 2156, "loss": 1.441, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.848677313755872e-05, "epoch": 0.22, "percentage": 11.13, "elapsed_time": "1:34:21", "remaining_time": "12:33:20"}
+{"current_steps": 250, "total_steps": 2156, "loss": 1.4496, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.835946975321647e-05, "epoch": 0.23, "percentage": 11.6, "elapsed_time": "1:38:16", "remaining_time": "12:29:16"}
+{"current_steps": 260, "total_steps": 2156, "loss": 1.4459, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.822720664465827e-05, "epoch": 0.24, "percentage": 12.06, "elapsed_time": "1:42:13", "remaining_time": "12:25:28"}
+{"current_steps": 270, "total_steps": 2156, "loss": 1.4458, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.809001189422287e-05, "epoch": 0.25, "percentage": 12.52, "elapsed_time": "1:46:20", "remaining_time": "12:22:48"}
+{"current_steps": 280, "total_steps": 2156, "loss": 1.4484, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.794791463134399e-05, "epoch": 0.26, "percentage": 12.99, "elapsed_time": "1:50:17", "remaining_time": "12:18:59"}
+{"current_steps": 290, "total_steps": 2156, "loss": 1.4541, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.780094502636552e-05, "epoch": 0.27, "percentage": 13.45, "elapsed_time": "1:54:05", "remaining_time": "12:14:08"}
+{"current_steps": 300, "total_steps": 2156, "loss": 1.437, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.764913428413572e-05, "epoch": 0.28, "percentage": 13.91, "elapsed_time": "1:58:02", "remaining_time": "12:10:16"}
+{"current_steps": 310, "total_steps": 2156, "loss": 1.4368, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.7492514637381727e-05, "epoch": 0.29, "percentage": 14.38, "elapsed_time": "2:01:56", "remaining_time": "12:06:09"}
+{"current_steps": 320, "total_steps": 2156, "loss": 1.4449, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.733111933986583e-05, "epoch": 0.3, "percentage": 14.84, "elapsed_time": "2:05:52", "remaining_time": "12:02:14"}
+{"current_steps": 330, "total_steps": 2156, "loss": 1.4256, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.716498265932501e-05, "epoch": 0.31, "percentage": 15.31, "elapsed_time": "2:09:46", "remaining_time": "11:58:04"}
+{"current_steps": 340, "total_steps": 2156, "loss": 1.4378, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.699413987019512e-05, "epoch": 0.32, "percentage": 15.77, "elapsed_time": "2:13:49", "remaining_time": "11:54:45"}
+{"current_steps": 350, "total_steps": 2156, "loss": 1.454, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.681862724612141e-05, "epoch": 0.32, "percentage": 16.23, "elapsed_time": "2:17:55", "remaining_time": "11:51:41"}
+{"current_steps": 360, "total_steps": 2156, "loss": 1.4207, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.663848205225674e-05, "epoch": 0.33, "percentage": 16.7, "elapsed_time": "2:21:56", "remaining_time": "11:48:08"}
+{"current_steps": 370, "total_steps": 2156, "loss": 1.4286, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.645374253734949e-05, "epoch": 0.34, "percentage": 17.16, "elapsed_time": "2:25:51", "remaining_time": "11:44:02"}
+{"current_steps": 380, "total_steps": 2156, "loss": 1.4244, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.626444792562244e-05, "epoch": 0.35, "percentage": 17.63, "elapsed_time": "2:29:47", "remaining_time": "11:40:03"}
+{"current_steps": 390, "total_steps": 2156, "loss": 1.4342, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.607063840844463e-05, "epoch": 0.36, "percentage": 18.09, "elapsed_time": "2:33:48", "remaining_time": "11:36:28"}
+{"current_steps": 400, "total_steps": 2156, "loss": 1.4241, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.587235513579791e-05, "epoch": 0.37, "percentage": 18.55, "elapsed_time": "2:37:54", "remaining_time": "11:33:12"}
+{"current_steps": 410, "total_steps": 2156, "loss": 1.443, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.5669640207539786e-05, "epoch": 0.38, "percentage": 19.02, "elapsed_time": "2:41:50", "remaining_time": "11:29:13"}
+{"current_steps": 420, "total_steps": 2156, "loss": 1.4281, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.546253666446484e-05, "epoch": 0.39, "percentage": 19.48, "elapsed_time": "2:45:55", "remaining_time": "11:25:47"}
+{"current_steps": 430, "total_steps": 2156, "loss": 1.4193, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.525108847916614e-05, "epoch": 0.4, "percentage": 19.94, "elapsed_time": "2:49:54", "remaining_time": "11:21:58"}
+{"current_steps": 440, "total_steps": 2156, "loss": 1.4154, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.503534054669892e-05, "epoch": 0.41, "percentage": 20.41, "elapsed_time": "2:53:52", "remaining_time": "11:18:07"}
+{"current_steps": 450, "total_steps": 2156, "loss": 1.4273, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.481533867504841e-05, "epoch": 0.42, "percentage": 20.87, "elapsed_time": "2:58:00", "remaining_time": "11:14:50"}
+{"current_steps": 460, "total_steps": 2156, "loss": 1.4329, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.4591129575403765e-05, "epoch": 0.43, "percentage": 21.34, "elapsed_time": "3:01:55", "remaining_time": "11:10:43"}
+{"current_steps": 470, "total_steps": 2156, "loss": 1.4102, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.43627608522403e-05, "epoch": 0.44, "percentage": 21.8, "elapsed_time": "3:05:54", "remaining_time": "11:06:53"}
+{"current_steps": 480, "total_steps": 2156, "loss": 1.4244, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.4130280993211974e-05, "epoch": 0.45, "percentage": 22.26, "elapsed_time": "3:09:56", "remaining_time": "11:03:12"}
+{"current_steps": 490, "total_steps": 2156, "loss": 1.4407, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.389373935885646e-05, "epoch": 0.45, "percentage": 22.73, "elapsed_time": "3:13:57", "remaining_time": "10:59:27"}
+{"current_steps": 500, "total_steps": 2156, "loss": 1.4282, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.365318617211479e-05, "epoch": 0.46, "percentage": 23.19, "elapsed_time": "3:17:54", "remaining_time": "10:55:29"}
+{"current_steps": 510, "total_steps": 2156, "loss": 1.4269, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.340867250766794e-05, "epoch": 0.47, "percentage": 23.65, "elapsed_time": "3:21:49", "remaining_time": "10:51:22"}
+{"current_steps": 520, "total_steps": 2156, "loss": 1.43, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.316025028109258e-05, "epoch": 0.48, "percentage": 24.12, "elapsed_time": "3:25:47", "remaining_time": "10:47:25"}
+{"current_steps": 530, "total_steps": 2156, "loss": 1.4271, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.2907972237838225e-05, "epoch": 0.49, "percentage": 24.58, "elapsed_time": "3:29:54", "remaining_time": "10:43:58"}
+{"current_steps": 540, "total_steps": 2156, "loss": 1.4321, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.2651891942028274e-05, "epoch": 0.5, "percentage": 25.05, "elapsed_time": "3:33:53", "remaining_time": "10:40:04"}
+{"current_steps": 550, "total_steps": 2156, "loss": 1.4083, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.239206376508717e-05, "epoch": 0.51, "percentage": 25.51, "elapsed_time": "3:37:48", "remaining_time": "10:36:00"}
+{"current_steps": 560, "total_steps": 2156, "loss": 1.4168, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.212854287419611e-05, "epoch": 0.52, "percentage": 25.97, "elapsed_time": "3:41:42", "remaining_time": "10:31:50"}
+{"current_steps": 570, "total_steps": 2156, "loss": 1.4144, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.1861385220579934e-05, "epoch": 0.53, "percentage": 26.44, "elapsed_time": "3:45:37", "remaining_time": "10:27:47"}
+{"current_steps": 580, "total_steps": 2156, "loss": 1.4106, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.1590647527627404e-05, "epoch": 0.54, "percentage": 26.9, "elapsed_time": "3:49:26", "remaining_time": "10:23:26"}
+{"current_steps": 590, "total_steps": 2156, "loss": 1.4202, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.131638727884762e-05, "epoch": 0.55, "percentage": 27.37, "elapsed_time": "3:53:19", "remaining_time": "10:19:19"}
+{"current_steps": 600, "total_steps": 2156, "loss": 1.417, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.103866270566498e-05, "epoch": 0.56, "percentage": 27.83, "elapsed_time": "3:57:15", "remaining_time": "10:15:18"}
+{"current_steps": 610, "total_steps": 2156, "loss": 1.4324, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.075753277505544e-05, "epoch": 0.57, "percentage": 28.29, "elapsed_time": "4:01:13", "remaining_time": "10:11:23"}
+{"current_steps": 620, "total_steps": 2156, "loss": 1.4147, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.0473057177026484e-05, "epoch": 0.57, "percentage": 28.76, "elapsed_time": "4:05:23", "remaining_time": "10:07:57"}
+{"current_steps": 630, "total_steps": 2156, "loss": 1.4269, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.018529631194369e-05, "epoch": 0.58, "percentage": 29.22, "elapsed_time": "4:09:20", "remaining_time": "10:03:58"}
+{"current_steps": 640, "total_steps": 2156, "loss": 1.4133, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.989431127770635e-05, "epoch": 0.59, "percentage": 29.68, "elapsed_time": "4:13:09", "remaining_time": "9:59:40"}
+{"current_steps": 650, "total_steps": 2156, "loss": 1.4173, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.960016385677513e-05, "epoch": 0.6, "percentage": 30.15, "elapsed_time": "4:17:06", "remaining_time": "9:55:41"}
+{"current_steps": 660, "total_steps": 2156, "loss": 1.4007, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.9302916503054246e-05, "epoch": 0.61, "percentage": 30.61, "elapsed_time": "4:21:12", "remaining_time": "9:52:04"}
+{"current_steps": 670, "total_steps": 2156, "loss": 1.4117, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.9002632328631164e-05, "epoch": 0.62, "percentage": 31.08, "elapsed_time": "4:25:07", "remaining_time": "9:48:00"}
+{"current_steps": 680, "total_steps": 2156, "loss": 1.4247, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.8699375090376534e-05, "epoch": 0.63, "percentage": 31.54, "elapsed_time": "4:29:03", "remaining_time": "9:44:01"}
+{"current_steps": 690, "total_steps": 2156, "loss": 1.4312, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.8393209176407223e-05, "epoch": 0.64, "percentage": 32.0, "elapsed_time": "4:32:59", "remaining_time": "9:40:00"}
+{"current_steps": 700, "total_steps": 2156, "loss": 1.4136, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.8084199592415305e-05, "epoch": 0.65, "percentage": 32.47, "elapsed_time": "4:37:02", "remaining_time": "9:36:13"}
+{"current_steps": 710, "total_steps": 2156, "loss": 1.4132, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.777241194786591e-05, "epoch": 0.66, "percentage": 32.93, "elapsed_time": "4:40:58", "remaining_time": "9:32:13"}
+{"current_steps": 720, "total_steps": 2156, "loss": 1.4145, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.745791244206697e-05, "epoch": 0.67, "percentage": 33.4, "elapsed_time": "4:44:52", "remaining_time": "9:28:10"}
+{"current_steps": 730, "total_steps": 2156, "loss": 1.3951, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.714076785011359e-05, "epoch": 0.68, "percentage": 33.86, "elapsed_time": "4:48:57", "remaining_time": "9:24:26"}
+{"current_steps": 740, "total_steps": 2156, "loss": 1.4139, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.682104550871031e-05, "epoch": 0.69, "percentage": 34.32, "elapsed_time": "4:52:56", "remaining_time": "9:20:32"}
+{"current_steps": 750, "total_steps": 2156, "loss": 1.4078, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.649881330187401e-05, "epoch": 0.7, "percentage": 34.79, "elapsed_time": "4:56:50", "remaining_time": "9:16:28"}
+{"current_steps": 760, "total_steps": 2156, "loss": 1.4048, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.617413964652067e-05, "epoch": 0.7, "percentage": 35.25, "elapsed_time": "5:00:52", "remaining_time": "9:12:40"}
+{"current_steps": 770, "total_steps": 2156, "loss": 1.4115, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.5847093477938956e-05, "epoch": 0.71, "percentage": 35.71, "elapsed_time": "5:04:55", "remaining_time": "9:08:51"}
+{"current_steps": 780, "total_steps": 2156, "loss": 1.4018, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.551774423515378e-05, "epoch": 0.72, "percentage": 36.18, "elapsed_time": "5:08:45", "remaining_time": "9:04:40"}
+{"current_steps": 790, "total_steps": 2156, "loss": 1.4091, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.518616184618288e-05, "epoch": 0.73, "percentage": 36.64, "elapsed_time": "5:12:43", "remaining_time": "9:00:44"}
+{"current_steps": 800, "total_steps": 2156, "loss": 1.4135, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.4852416713189526e-05, "epoch": 0.74, "percentage": 37.11, "elapsed_time": "5:16:40", "remaining_time": "8:56:45"}
+{"current_steps": 810, "total_steps": 2156, "loss": 1.3977, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.4516579697534705e-05, "epoch": 0.75, "percentage": 37.57, "elapsed_time": "5:20:49", "remaining_time": "8:53:07"}
+{"current_steps": 820, "total_steps": 2156, "loss": 1.4123, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.417872210473162e-05, "epoch": 0.76, "percentage": 38.03, "elapsed_time": "5:24:55", "remaining_time": "8:49:23"}
+{"current_steps": 830, "total_steps": 2156, "loss": 1.4096, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.3838915669306034e-05, "epoch": 0.77, "percentage": 38.5, "elapsed_time": "5:28:51", "remaining_time": "8:45:22"}
+{"current_steps": 840, "total_steps": 2156, "loss": 1.4031, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.349723253956542e-05, "epoch": 0.78, "percentage": 38.96, "elapsed_time": "5:32:48", "remaining_time": "8:41:24"}
+{"current_steps": 850, "total_steps": 2156, "loss": 1.4104, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.315374526228036e-05, "epoch": 0.79, "percentage": 39.42, "elapsed_time": "5:36:46", "remaining_time": "8:37:26"}
+{"current_steps": 860, "total_steps": 2156, "loss": 1.3972, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.2808526767281225e-05, "epoch": 0.8, "percentage": 39.89, "elapsed_time": "5:40:47", "remaining_time": "8:33:33"}
+{"current_steps": 870, "total_steps": 2156, "loss": 1.3987, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.246165035197364e-05, "epoch": 0.81, "percentage": 40.35, "elapsed_time": "5:44:47", "remaining_time": "8:29:38"}
+{"current_steps": 880, "total_steps": 2156, "loss": 1.4197, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.211318966577581e-05, "epoch": 0.82, "percentage": 40.82, "elapsed_time": "5:48:49", "remaining_time": "8:25:47"}
+{"current_steps": 890, "total_steps": 2156, "loss": 1.4169, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.176321869448116e-05, "epoch": 0.83, "percentage": 41.28, "elapsed_time": "5:52:46", "remaining_time": "8:21:48"}
+{"current_steps": 900, "total_steps": 2156, "loss": 1.4055, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.1411811744549536e-05, "epoch": 0.83, "percentage": 41.74, "elapsed_time": "5:56:38", "remaining_time": "8:17:42"}
+{"current_steps": 910, "total_steps": 2156, "loss": 1.402, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.105904342733032e-05, "epoch": 0.84, "percentage": 42.21, "elapsed_time": "6:00:35", "remaining_time": "8:13:44"}
+{"current_steps": 920, "total_steps": 2156, "loss": 1.4148, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.070498864322081e-05, "epoch": 0.85, "percentage": 42.67, "elapsed_time": "6:04:30", "remaining_time": "8:09:42"}
+{"current_steps": 930, "total_steps": 2156, "loss": 1.4007, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.034972256576328e-05, "epoch": 0.86, "percentage": 43.14, "elapsed_time": "6:08:25", "remaining_time": "8:05:41"}
+{"current_steps": 940, "total_steps": 2156, "loss": 1.3926, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.999332062568395e-05, "epoch": 0.87, "percentage": 43.6, "elapsed_time": "6:12:19", "remaining_time": "8:01:39"}
+{"current_steps": 950, "total_steps": 2156, "loss": 1.4099, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.9635858494877384e-05, "epoch": 0.88, "percentage": 44.06, "elapsed_time": "6:16:19", "remaining_time": "7:57:44"}
+{"current_steps": 960, "total_steps": 2156, "loss": 1.401, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.9277412070339782e-05, "epoch": 0.89, "percentage": 44.53, "elapsed_time": "6:20:23", "remaining_time": "7:53:53"}
+{"current_steps": 970, "total_steps": 2156, "loss": 1.3995, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.891805745805429e-05, "epoch": 0.9, "percentage": 44.99, "elapsed_time": "6:24:32", "remaining_time": "7:50:09"}
+{"current_steps": 980, "total_steps": 2156, "loss": 1.4108, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.8557870956832132e-05, "epoch": 0.91, "percentage": 45.45, "elapsed_time": "6:28:27", "remaining_time": "7:46:09"}
+{"current_steps": 990, "total_steps": 2156, "loss": 1.4006, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.8196929042112652e-05, "epoch": 0.92, "percentage": 45.92, "elapsed_time": "6:32:21", "remaining_time": "7:42:07"}
+{"current_steps": 1000, "total_steps": 2156, "loss": 1.4045, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.783530834972594e-05, "epoch": 0.93, "percentage": 46.38, "elapsed_time": "6:36:20", "remaining_time": "7:38:09"}
+{"current_steps": 1010, "total_steps": 2156, "loss": 1.3907, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.7473085659621377e-05, "epoch": 0.94, "percentage": 46.85, "elapsed_time": "6:40:20", "remaining_time": "7:34:14"}
+{"current_steps": 1020, "total_steps": 2156, "loss": 1.3872, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.711033787956555e-05, "epoch": 0.95, "percentage": 47.31, "elapsed_time": "6:44:22", "remaining_time": "7:30:21"}
+{"current_steps": 1030, "total_steps": 2156, "loss": 1.405, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.6747142028813105e-05, "epoch": 0.96, "percentage": 47.77, "elapsed_time": "6:48:20", "remaining_time": "7:26:24"}
+{"current_steps": 1040, "total_steps": 2156, "loss": 1.3952, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.638357522175383e-05, "epoch": 0.96, "percentage": 48.24, "elapsed_time": "6:52:13", "remaining_time": "7:22:20"}
+{"current_steps": 1050, "total_steps": 2156, "loss": 1.3901, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.6019714651539646e-05, "epoch": 0.97, "percentage": 48.7, "elapsed_time": "6:56:14", "remaining_time": "7:18:26"}
+{"current_steps": 1060, "total_steps": 2156, "loss": 1.4027, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.565563757369475e-05, "epoch": 0.98, "percentage": 49.17, "elapsed_time": "7:00:12", "remaining_time": "7:14:29"}
+{"current_steps": 1070, "total_steps": 2156, "loss": 1.3909, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.529142128971268e-05, "epoch": 0.99, "percentage": 49.63, "elapsed_time": "7:04:17", "remaining_time": "7:10:38"}
+{"current_steps": 1080, "total_steps": 2156, "loss": 1.4039, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.492714313064342e-05, "epoch": 1.0, "percentage": 50.09, "elapsed_time": "7:08:25", "remaining_time": "7:06:50"}
+{"current_steps": 1090, "total_steps": 2156, "loss": 1.3964, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.45628804406744e-05, "epoch": 1.01, "percentage": 50.56, "elapsed_time": "7:12:23", "remaining_time": "7:02:52"}
+{"current_steps": 1100, "total_steps": 2156, "loss": 1.4076, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.419871056070862e-05, "epoch": 1.02, "percentage": 51.02, "elapsed_time": "7:16:17", "remaining_time": "6:58:50"}
+{"current_steps": 1110, "total_steps": 2156, "loss": 1.3891, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.3834710811943514e-05, "epoch": 1.03, "percentage": 51.48, "elapsed_time": "7:20:12", "remaining_time": "6:54:50"}
+{"current_steps": 1120, "total_steps": 2156, "loss": 1.4004, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.3470958479453938e-05, "epoch": 1.04, "percentage": 51.95, "elapsed_time": "7:24:12", "remaining_time": "6:50:53"}
+{"current_steps": 1130, "total_steps": 2156, "loss": 1.3879, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.3107530795782877e-05, "epoch": 1.05, "percentage": 52.41, "elapsed_time": "7:28:14", "remaining_time": "6:46:58"}
+{"current_steps": 1140, "total_steps": 2156, "loss": 1.3879, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.2744504924543313e-05, "epoch": 1.06, "percentage": 52.88, "elapsed_time": "7:32:09", "remaining_time": "6:42:58"}
+{"current_steps": 1150, "total_steps": 2156, "loss": 1.3851, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.23819579440347e-05, "epoch": 1.07, "percentage": 53.34, "elapsed_time": "7:36:09", "remaining_time": "6:39:02"}
+{"current_steps": 1160, "total_steps": 2156, "loss": 1.3846, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.2019966830877545e-05, "epoch": 1.08, "percentage": 53.8, "elapsed_time": "7:40:18", "remaining_time": "6:35:13"}
+{"current_steps": 1170, "total_steps": 2156, "loss": 1.3936, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.1658608443669635e-05, "epoch": 1.08, "percentage": 54.27, "elapsed_time": "7:44:16", "remaining_time": "6:31:15"}
+{"current_steps": 1180, "total_steps": 2156, "loss": 1.3841, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.1297959506667224e-05, "epoch": 1.09, "percentage": 54.73, "elapsed_time": "7:48:20", "remaining_time": "6:27:22"}
+{"current_steps": 1190, "total_steps": 2156, "loss": 1.3969, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.0938096593494855e-05, "epoch": 1.1, "percentage": 55.19, "elapsed_time": "7:52:27", "remaining_time": "6:23:31"}
+{"current_steps": 1200, "total_steps": 2156, "loss": 1.3953, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.057909611088709e-05, "epoch": 1.11, "percentage": 55.66, "elapsed_time": "7:56:26", "remaining_time": "6:19:34"}
+{"current_steps": 1210, "total_steps": 2156, "loss": 1.3951, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.02210342824657e-05, "epoch": 1.12, "percentage": 56.12, "elapsed_time": "8:00:22", "remaining_time": "6:15:33"}
+{"current_steps": 1220, "total_steps": 2156, "loss": 1.4012, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.9863987132555706e-05, "epoch": 1.13, "percentage": 56.59, "elapsed_time": "8:04:18", "remaining_time": "6:11:34"}
+{"current_steps": 1230, "total_steps": 2156, "loss": 1.3998, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.9508030470043806e-05, "epoch": 1.14, "percentage": 57.05, "elapsed_time": "8:08:21", "remaining_time": "6:07:39"}
+{"current_steps": 1240, "total_steps": 2156, "loss": 1.3933, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.915323987228247e-05, "epoch": 1.15, "percentage": 57.51, "elapsed_time": "8:12:22", "remaining_time": "6:03:43"}
+{"current_steps": 1250, "total_steps": 2156, "loss": 1.3977, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.8799690669043212e-05, "epoch": 1.16, "percentage": 57.98, "elapsed_time": "8:16:18", "remaining_time": "5:59:43"}
+{"current_steps": 1260, "total_steps": 2156, "loss": 1.3818, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.8447457926522454e-05, "epoch": 1.17, "percentage": 58.44, "elapsed_time": "8:20:15", "remaining_time": "5:55:44"}
+{"current_steps": 1270, "total_steps": 2156, "loss": 1.4032, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.8096616431403325e-05, "epoch": 1.18, "percentage": 58.91, "elapsed_time": "8:24:17", "remaining_time": "5:51:48"}
+{"current_steps": 1280, "total_steps": 2156, "loss": 1.3853, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.7747240674976857e-05, "epoch": 1.19, "percentage": 59.37, "elapsed_time": "8:28:14", "remaining_time": "5:47:49"}
+{"current_steps": 1290, "total_steps": 2156, "loss": 1.3988, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.7399404837325796e-05, "epoch": 1.2, "percentage": 59.83, "elapsed_time": "8:32:08", "remaining_time": "5:43:48"}
+{"current_steps": 1300, "total_steps": 2156, "loss": 1.3963, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.7053182771574633e-05, "epoch": 1.21, "percentage": 60.3, "elapsed_time": "8:35:59", "remaining_time": "5:39:45"}
+{"current_steps": 1310, "total_steps": 2156, "loss": 1.3881, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.6708647988208887e-05, "epoch": 1.21, "percentage": 60.76, "elapsed_time": "8:39:58", "remaining_time": "5:35:47"}
+{"current_steps": 1320, "total_steps": 2156, "loss": 1.3804, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.6365873639467315e-05, "epoch": 1.22, "percentage": 61.22, "elapsed_time": "8:43:59", "remaining_time": "5:31:51"}
+{"current_steps": 1330, "total_steps": 2156, "loss": 1.3952, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.602493250381003e-05, "epoch": 1.23, "percentage": 61.69, "elapsed_time": "8:48:00", "remaining_time": "5:27:55"}
+{"current_steps": 1340, "total_steps": 2156, "loss": 1.4015, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.5685896970466123e-05, "epoch": 1.24, "percentage": 62.15, "elapsed_time": "8:51:53", "remaining_time": "5:23:53"}
+{"current_steps": 1350, "total_steps": 2156, "loss": 1.3797, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.534883902406375e-05, "epoch": 1.25, "percentage": 62.62, "elapsed_time": "8:55:45", "remaining_time": "5:19:51"}
+{"current_steps": 1360, "total_steps": 2156, "loss": 1.3931, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.5013830229346326e-05, "epoch": 1.26, "percentage": 63.08, "elapsed_time": "8:59:40", "remaining_time": "5:15:52"}
+{"current_steps": 1370, "total_steps": 2156, "loss": 1.3946, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.4680941715977722e-05, "epoch": 1.27, "percentage": 63.54, "elapsed_time": "9:03:37", "remaining_time": "5:11:53"}
+{"current_steps": 1380, "total_steps": 2156, "loss": 1.3851, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.4350244163439892e-05, "epoch": 1.28, "percentage": 64.01, "elapsed_time": "9:07:33", "remaining_time": "5:07:54"}
+{"current_steps": 1390, "total_steps": 2156, "loss": 1.3891, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.4021807786026108e-05, "epoch": 1.29, "percentage": 64.47, "elapsed_time": "9:11:38", "remaining_time": "5:03:59"}
+{"current_steps": 1400, "total_steps": 2156, "loss": 1.3958, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.3695702317932862e-05, "epoch": 1.3, "percentage": 64.94, "elapsed_time": "9:15:33", "remaining_time": "4:59:59"}
+{"current_steps": 1410, "total_steps": 2156, "loss": 1.3909, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.337199699845387e-05, "epoch": 1.31, "percentage": 65.4, "elapsed_time": "9:19:20", "remaining_time": "4:55:56"}
+{"current_steps": 1420, "total_steps": 2156, "loss": 1.3785, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.3050760557279023e-05, "epoch": 1.32, "percentage": 65.86, "elapsed_time": "9:23:32", "remaining_time": "4:52:05"}
+{"current_steps": 1430, "total_steps": 2156, "loss": 1.3882, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.2732061199901562e-05, "epoch": 1.33, "percentage": 66.33, "elapsed_time": "9:27:34", "remaining_time": "4:48:09"}
+{"current_steps": 1440, "total_steps": 2156, "loss": 1.3872, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.2415966593136547e-05, "epoch": 1.34, "percentage": 66.79, "elapsed_time": "9:31:33", "remaining_time": "4:44:11"}
+{"current_steps": 1450, "total_steps": 2156, "loss": 1.4032, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.2102543850753808e-05, "epoch": 1.34, "percentage": 67.25, "elapsed_time": "9:35:31", "remaining_time": "4:40:13"}
+{"current_steps": 1460, "total_steps": 2156, "loss": 1.3928, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.1791859519228138e-05, "epoch": 1.35, "percentage": 67.72, "elapsed_time": "9:39:28", "remaining_time": "4:36:14"}
+{"current_steps": 1470, "total_steps": 2156, "loss": 1.3785, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.148397956361007e-05, "epoch": 1.36, "percentage": 68.18, "elapsed_time": "9:43:33", "remaining_time": "4:32:19"}
+{"current_steps": 1480, "total_steps": 2156, "loss": 1.3797, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.1178969353520018e-05, "epoch": 1.37, "percentage": 68.65, "elapsed_time": "9:47:27", "remaining_time": "4:28:19"}
+{"current_steps": 1490, "total_steps": 2156, "loss": 1.3966, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.087689364926897e-05, "epoch": 1.38, "percentage": 69.11, "elapsed_time": "9:51:30", "remaining_time": "4:24:23"}
+{"current_steps": 1500, "total_steps": 2156, "loss": 1.3849, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.0577816588108378e-05, "epoch": 1.39, "percentage": 69.57, "elapsed_time": "9:55:21", "remaining_time": "4:20:22"}
+{"current_steps": 1510, "total_steps": 2156, "loss": 1.3732, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.0281801670612449e-05, "epoch": 1.4, "percentage": 70.04, "elapsed_time": "9:59:19", "remaining_time": "4:16:23"}
+{"current_steps": 1520, "total_steps": 2156, "loss": 1.4018, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.988911747195603e-06, "epoch": 1.41, "percentage": 70.5, "elapsed_time": "10:03:22", "remaining_time": "4:12:27"}
+{"current_steps": 1530, "total_steps": 2156, "loss": 1.3902, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.699209004767953e-06, "epoch": 1.42, "percentage": 70.96, "elapsed_time": "10:07:13", "remaining_time": "4:08:26"}
+{"current_steps": 1540, "total_steps": 2156, "loss": 1.3991, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.412754953531663e-06, "epoch": 1.43, "percentage": 71.43, "elapsed_time": "10:11:04", "remaining_time": "4:04:25"}
+{"current_steps": 1550, "total_steps": 2156, "loss": 1.3824, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.129610413921005e-06, "epoch": 1.44, "percentage": 71.89, "elapsed_time": "10:14:57", "remaining_time": "4:00:25"}
+{"current_steps": 1560, "total_steps": 2156, "loss": 1.3904, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 8.849835503688847e-06, "epoch": 1.45, "percentage": 72.36, "elapsed_time": "10:18:49", "remaining_time": "3:56:25"}
+{"current_steps": 1570, "total_steps": 2156, "loss": 1.3937, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 8.57348962514234e-06, "epoch": 1.46, "percentage": 72.82, "elapsed_time": "10:22:47", "remaining_time": "3:52:27"}
+{"current_steps": 1580, "total_steps": 2156, "loss": 1.39, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 8.30063145253053e-06, "epoch": 1.46, "percentage": 73.28, "elapsed_time": "10:26:57", "remaining_time": "3:48:33"}
+{"current_steps": 1590, "total_steps": 2156, "loss": 1.3773, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 8.031318919586523e-06, "epoch": 1.47, "percentage": 73.75, "elapsed_time": "10:30:54", "remaining_time": "3:44:35"}
+{"current_steps": 1600, "total_steps": 2156, "loss": 1.3875, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 7.7656092072269e-06, "epoch": 1.48, "percentage": 74.21, "elapsed_time": "10:34:53", "remaining_time": "3:40:37"}
+{"current_steps": 1610, "total_steps": 2156, "loss": 1.3859, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 7.503558731410959e-06, "epoch": 1.49, "percentage": 74.68, "elapsed_time": "10:38:44", "remaining_time": "3:36:37"}
+{"current_steps": 1620, "total_steps": 2156, "loss": 1.3716, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 7.245223131162376e-06, "epoch": 1.5, "percentage": 75.14, "elapsed_time": "10:42:38", "remaining_time": "3:32:37"}
+{"current_steps": 1630, "total_steps": 2156, "loss": 1.3846, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.9906572567558285e-06, "epoch": 1.51, "percentage": 75.6, "elapsed_time": "10:46:30", "remaining_time": "3:28:37"}
+{"current_steps": 1640, "total_steps": 2156, "loss": 1.3741, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.739915158071106e-06, "epoch": 1.52, "percentage": 76.07, "elapsed_time": "10:50:24", "remaining_time": "3:24:38"}
+{"current_steps": 1650, "total_steps": 2156, "loss": 1.3827, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.493050073117116e-06, "epoch": 1.53, "percentage": 76.53, "elapsed_time": "10:54:22", "remaining_time": "3:20:40"}
+{"current_steps": 1660, "total_steps": 2156, "loss": 1.3951, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.250114416728298e-06, "epoch": 1.54, "percentage": 76.99, "elapsed_time": "10:58:20", "remaining_time": "3:16:42"}
+{"current_steps": 1670, "total_steps": 2156, "loss": 1.3875, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.011159769435823e-06, "epoch": 1.55, "percentage": 77.46, "elapsed_time": "11:02:18", "remaining_time": "3:12:44"}
+{"current_steps": 1680, "total_steps": 2156, "loss": 1.3783, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.776236866515947e-06, "epoch": 1.56, "percentage": 77.92, "elapsed_time": "11:06:17", "remaining_time": "3:08:46"}
+{"current_steps": 1690, "total_steps": 2156, "loss": 1.3888, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.545395587217763e-06, "epoch": 1.57, "percentage": 78.39, "elapsed_time": "11:10:20", "remaining_time": "3:04:50"}
+{"current_steps": 1700, "total_steps": 2156, "loss": 1.4006, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.318684944172752e-06, "epoch": 1.58, "percentage": 78.85, "elapsed_time": "11:14:18", "remaining_time": "3:00:52"}
+{"current_steps": 1710, "total_steps": 2156, "loss": 1.4041, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.096153072988366e-06, "epoch": 1.59, "percentage": 79.31, "elapsed_time": "11:18:08", "remaining_time": "2:56:52"}
+{"current_steps": 1720, "total_steps": 2156, "loss": 1.3865, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.8778472220277874e-06, "epoch": 1.59, "percentage": 79.78, "elapsed_time": "11:21:59", "remaining_time": "2:52:52"}
+{"current_steps": 1730, "total_steps": 2156, "loss": 1.3789, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.6638137423780395e-06, "epoch": 1.6, "percentage": 80.24, "elapsed_time": "11:26:02", "remaining_time": "2:48:55"}
+{"current_steps": 1740, "total_steps": 2156, "loss": 1.3861, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.454098078008667e-06, "epoch": 1.61, "percentage": 80.71, "elapsed_time": "11:30:04", "remaining_time": "2:44:59"}
+{"current_steps": 1750, "total_steps": 2156, "loss": 1.3803, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.248744756122986e-06, "epoch": 1.62, "percentage": 81.17, "elapsed_time": "11:34:05", "remaining_time": "2:41:01"}
+{"current_steps": 1760, "total_steps": 2156, "loss": 1.3695, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.047797377703985e-06, "epoch": 1.63, "percentage": 81.63, "elapsed_time": "11:38:04", "remaining_time": "2:37:04"}
+{"current_steps": 1770, "total_steps": 2156, "loss": 1.3937, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.851298608256892e-06, "epoch": 1.64, "percentage": 82.1, "elapsed_time": "11:42:06", "remaining_time": "2:33:06"}
+{"current_steps": 1780, "total_steps": 2156, "loss": 1.3865, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.6592901687503566e-06, "epoch": 1.65, "percentage": 82.56, "elapsed_time": "11:46:05", "remaining_time": "2:29:09"}
+{"current_steps": 1790, "total_steps": 2156, "loss": 1.3799, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.471812826758178e-06, "epoch": 1.66, "percentage": 83.02, "elapsed_time": "11:50:08", "remaining_time": "2:25:12"}
+{"current_steps": 1800, "total_steps": 2156, "loss": 1.3911, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.288906387803464e-06, "epoch": 1.67, "percentage": 83.49, "elapsed_time": "11:54:06", "remaining_time": "2:21:14"}
+{"current_steps": 1810, "total_steps": 2156, "loss": 1.3801, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.1106096869070483e-06, "epoch": 1.68, "percentage": 83.95, "elapsed_time": "11:58:01", "remaining_time": "2:17:15"}
+{"current_steps": 1820, "total_steps": 2156, "loss": 1.382, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.9369605803419715e-06, "epoch": 1.69, "percentage": 84.42, "elapsed_time": "12:02:04", "remaining_time": "2:13:18"}
+{"current_steps": 1830, "total_steps": 2156, "loss": 1.3898, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.767995937595766e-06, "epoch": 1.7, "percentage": 84.88, "elapsed_time": "12:06:04", "remaining_time": "2:09:20"}
+{"current_steps": 1840, "total_steps": 2156, "loss": 1.3752, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.6037516335422728e-06, "epoch": 1.71, "percentage": 85.34, "elapsed_time": "12:10:12", "remaining_time": "2:05:24"}
+{"current_steps": 1850, "total_steps": 2156, "loss": 1.3855, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.4442625408246074e-06, "epoch": 1.72, "percentage": 85.81, "elapsed_time": "12:14:13", "remaining_time": "2:01:26"}
+{"current_steps": 1860, "total_steps": 2156, "loss": 1.3665, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.289562522450947e-06, "epoch": 1.72, "percentage": 86.27, "elapsed_time": "12:18:04", "remaining_time": "1:57:27"}
+{"current_steps": 1870, "total_steps": 2156, "loss": 1.3747, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.1396844246046903e-06, "epoch": 1.73, "percentage": 86.73, "elapsed_time": "12:22:07", "remaining_time": "1:53:30"}
+{"current_steps": 1880, "total_steps": 2156, "loss": 1.3861, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.9946600696704592e-06, "epoch": 1.74, "percentage": 87.2, "elapsed_time": "12:25:58", "remaining_time": "1:49:30"}
+{"current_steps": 1890, "total_steps": 2156, "loss": 1.3844, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.8545202494775509e-06, "epoch": 1.75, "percentage": 87.66, "elapsed_time": "12:29:57", "remaining_time": "1:45:32"}
+{"current_steps": 1900, "total_steps": 2156, "loss": 1.3796, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.7192947187621434e-06, "epoch": 1.76, "percentage": 88.13, "elapsed_time": "12:33:59", "remaining_time": "1:41:35"}
+{"current_steps": 1910, "total_steps": 2156, "loss": 1.3963, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.5890121888497366e-06, "epoch": 1.77, "percentage": 88.59, "elapsed_time": "12:37:55", "remaining_time": "1:37:37"}
+{"current_steps": 1920, "total_steps": 2156, "loss": 1.385, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.463700321559075e-06, "epoch": 1.78, "percentage": 89.05, "elapsed_time": "12:41:56", "remaining_time": "1:33:39"}
+{"current_steps": 1930, "total_steps": 2156, "loss": 1.3811, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.3433857233289714e-06, "epoch": 1.79, "percentage": 89.52, "elapsed_time": "12:45:53", "remaining_time": "1:29:41"}
+{"current_steps": 1940, "total_steps": 2156, "loss": 1.384, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.2280939395691859e-06, "epoch": 1.8, "percentage": 89.98, "elapsed_time": "12:49:48", "remaining_time": "1:25:42"}
+{"current_steps": 1950, "total_steps": 2156, "loss": 1.3821, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.1178494492365465e-06, "epoch": 1.81, "percentage": 90.45, "elapsed_time": "12:53:43", "remaining_time": "1:21:44"}
+{"current_steps": 1960, "total_steps": 2156, "loss": 1.3978, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.0126756596375686e-06, "epoch": 1.82, "percentage": 90.91, "elapsed_time": "12:57:51", "remaining_time": "1:17:47"}
+{"current_steps": 1970, "total_steps": 2156, "loss": 1.3724, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.125949014585383e-07, "epoch": 1.83, "percentage": 91.37, "elapsed_time": "13:02:04", "remaining_time": "1:13:50"}
+{"current_steps": 1980, "total_steps": 2156, "loss": 1.3851, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 8.176284240242638e-07, "epoch": 1.84, "percentage": 91.84, "elapsed_time": "13:06:07", "remaining_time": "1:09:52"}
+{"current_steps": 1990, "total_steps": 2156, "loss": 1.3793, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 7.277963907863478e-07, "epoch": 1.85, "percentage": 92.3, "elapsed_time": "13:10:10", "remaining_time": "1:05:54"}
+{"current_steps": 2000, "total_steps": 2156, "loss": 1.3946, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.431178750420513e-07, "epoch": 1.85, "percentage": 92.76, "elapsed_time": "13:14:06", "remaining_time": "1:01:56"}
+{"current_steps": 2010, "total_steps": 2156, "loss": 1.3794, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.636108558846359e-07, "epoch": 1.86, "percentage": 93.23, "elapsed_time": "13:18:10", "remaining_time": "0:57:58"}
+{"current_steps": 2020, "total_steps": 2156, "loss": 1.3886, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.892922143859918e-07, "epoch": 1.87, "percentage": 93.69, "elapsed_time": "13:22:06", "remaining_time": "0:54:00"}
+{"current_steps": 2030, "total_steps": 2156, "loss": 1.3794, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.201777300124249e-07, "epoch": 1.88, "percentage": 94.16, "elapsed_time": "13:26:06", "remaining_time": "0:50:02"}
+{"current_steps": 2040, "total_steps": 2156, "loss": 1.3828, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.562820772743392e-07, "epoch": 1.89, "percentage": 94.62, "elapsed_time": "13:30:14", "remaining_time": "0:46:04"}
+{"current_steps": 2050, "total_steps": 2156, "loss": 1.374, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.9761882261050143e-07, "epoch": 1.9, "percentage": 95.08, "elapsed_time": "13:34:17", "remaining_time": "0:42:06"}
+{"current_steps": 2060, "total_steps": 2156, "loss": 1.3804, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.4420042150761215e-07, "epoch": 1.91, "percentage": 95.55, "elapsed_time": "13:38:15", "remaining_time": "0:38:07"}
+{"current_steps": 2070, "total_steps": 2156, "loss": 1.3772, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.9603821585572058e-07, "epoch": 1.92, "percentage": 96.01, "elapsed_time": "13:42:17", "remaining_time": "0:34:09"}
+{"current_steps": 2080, "total_steps": 2156, "loss": 1.3913, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.531424315400931e-07, "epoch": 1.93, "percentage": 96.47, "elapsed_time": "13:46:29", "remaining_time": "0:30:11"}
+{"current_steps": 2090, "total_steps": 2156, "loss": 1.3859, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.1552217627004425e-07, "epoch": 1.94, "percentage": 96.94, "elapsed_time": "13:50:32", "remaining_time": "0:26:13"}
+{"current_steps": 2100, "total_steps": 2156, "loss": 1.4035, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 8.318543764516961e-08, "epoch": 1.95, "percentage": 97.4, "elapsed_time": "13:54:40", "remaining_time": "0:22:15"}
+{"current_steps": 2110, "total_steps": 2156, "loss": 1.4047, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.613908145939428e-08, "epoch": 1.96, "percentage": 97.87, "elapsed_time": "13:58:37", "remaining_time": "0:18:16"}
+{"current_steps": 2120, "total_steps": 2156, "loss": 1.3886, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.438885024322769e-08, "epoch": 1.97, "percentage": 98.33, "elapsed_time": "14:02:39", "remaining_time": "0:14:18"}
+{"current_steps": 2130, "total_steps": 2156, "loss": 1.3874, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.7939362044494534e-08, "epoch": 1.97, "percentage": 98.79, "elapsed_time": "14:06:37", "remaining_time": "0:10:20"}
+{"current_steps": 2140, "total_steps": 2156, "loss": 1.3852, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.794109447824659e-09, "epoch": 1.98, "percentage": 99.26, "elapsed_time": "14:10:36", "remaining_time": "0:06:21"}
+{"current_steps": 2150, "total_steps": 2156, "loss": 1.3859, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.554588330934478e-10, "epoch": 1.99, "percentage": 99.72, "elapsed_time": "14:14:42", "remaining_time": "0:02:23"}
+{"current_steps": 2156, "total_steps": 2156, "loss": null, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 2.0, "percentage": 100.0, "elapsed_time": "14:17:05", "remaining_time": "0:00:00"}

trainer_state.json

@@ -0,0 +1,1315 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.999072786277237,
+  "global_step": 2156,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.999734597774032e-05,
+      "loss": 1.7592,
+      "step": 10
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.998938447446803e-05,
+      "loss": 1.6546,
+      "step": 20
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 4.997611718058365e-05,
+      "loss": 1.5632,
+      "step": 30
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.9957546913022665e-05,
+      "loss": 1.5245,
+      "step": 40
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 4.993367761465736e-05,
+      "loss": 1.5049,
+      "step": 50
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.9904514353459654e-05,
+      "loss": 1.492,
+      "step": 60
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.9870063321425105e-05,
+      "loss": 1.4803,
+      "step": 70
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 4.983033183325818e-05,
+      "loss": 1.4827,
+      "step": 80
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 4.97853283248192e-05,
+      "loss": 1.4866,
+      "step": 90
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 4.973506235133323e-05,
+      "loss": 1.486,
+      "step": 100
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 4.967954458536126e-05,
+      "loss": 1.4822,
+      "step": 110
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 4.9618786814534226e-05,
+      "loss": 1.4801,
+      "step": 120
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 4.955280193905022e-05,
+      "loss": 1.4764,
+      "step": 130
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.948160396893553e-05,
+      "loss": 1.4662,
+      "step": 140
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 4.9405208021069946e-05,
+      "loss": 1.4627,
+      "step": 150
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 4.9323630315977156e-05,
+      "loss": 1.4543,
+      "step": 160
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.9236888174380784e-05,
+      "loss": 1.4508,
+      "step": 170
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 4.91450000135268e-05,
+      "loss": 1.464,
+      "step": 180
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.9047985343273154e-05,
+      "loss": 1.4705,
+      "step": 190
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.894586476194739e-05,
+      "loss": 1.4588,
+      "step": 200
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.883865995197319e-05,
+      "loss": 1.4532,
+      "step": 210
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.8726393675266716e-05,
+      "loss": 1.4629,
+      "step": 220
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 4.860908976840376e-05,
+      "loss": 1.4429,
+      "step": 230
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 4.848677313755872e-05,
+      "loss": 1.441,
+      "step": 240
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 4.835946975321647e-05,
+      "loss": 1.4496,
+      "step": 250
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 4.822720664465827e-05,
+      "loss": 1.4459,
+      "step": 260
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 4.809001189422287e-05,
+      "loss": 1.4458,
+      "step": 270
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 4.794791463134399e-05,
+      "loss": 1.4484,
+      "step": 280
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 4.780094502636552e-05,
+      "loss": 1.4541,
+      "step": 290
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 4.764913428413572e-05,
+      "loss": 1.437,
+      "step": 300
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 4.7492514637381727e-05,
+      "loss": 1.4368,
+      "step": 310
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 4.733111933986583e-05,
+      "loss": 1.4449,
+      "step": 320
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 4.716498265932501e-05,
+      "loss": 1.4256,
+      "step": 330
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 4.699413987019512e-05,
+      "loss": 1.4378,
+      "step": 340
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 4.681862724612141e-05,
+      "loss": 1.454,
+      "step": 350
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 4.663848205225674e-05,
+      "loss": 1.4207,
+      "step": 360
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 4.645374253734949e-05,
+      "loss": 1.4286,
+      "step": 370
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 4.626444792562244e-05,
+      "loss": 1.4244,
+      "step": 380
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 4.607063840844463e-05,
+      "loss": 1.4342,
+      "step": 390
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 4.587235513579791e-05,
+      "loss": 1.4241,
+      "step": 400
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 4.5669640207539786e-05,
+      "loss": 1.443,
+      "step": 410
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 4.546253666446484e-05,
+      "loss": 1.4281,
+      "step": 420
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 4.525108847916614e-05,
+      "loss": 1.4193,
+      "step": 430
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 4.503534054669892e-05,
+      "loss": 1.4154,
+      "step": 440
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 4.481533867504841e-05,
+      "loss": 1.4273,
+      "step": 450
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 4.4591129575403765e-05,
+      "loss": 1.4329,
+      "step": 460
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 4.43627608522403e-05,
+      "loss": 1.4102,
+      "step": 470
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 4.4130280993211974e-05,
+      "loss": 1.4244,
+      "step": 480
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 4.389373935885646e-05,
+      "loss": 1.4407,
+      "step": 490
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 4.365318617211479e-05,
+      "loss": 1.4282,
+      "step": 500
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 4.340867250766794e-05,
+      "loss": 1.4269,
+      "step": 510
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 4.316025028109258e-05,
+      "loss": 1.43,
+      "step": 520
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 4.2907972237838225e-05,
+      "loss": 1.4271,
+      "step": 530
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 4.2651891942028274e-05,
+      "loss": 1.4321,
+      "step": 540
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 4.239206376508717e-05,
+      "loss": 1.4083,
+      "step": 550
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 4.212854287419611e-05,
+      "loss": 1.4168,
+      "step": 560
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 4.1861385220579934e-05,
+      "loss": 1.4144,
+      "step": 570
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 4.1590647527627404e-05,
+      "loss": 1.4106,
+      "step": 580
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 4.131638727884762e-05,
+      "loss": 1.4202,
+      "step": 590
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 4.103866270566498e-05,
+      "loss": 1.417,
+      "step": 600
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 4.075753277505544e-05,
+      "loss": 1.4324,
+      "step": 610
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 4.0473057177026484e-05,
+      "loss": 1.4147,
+      "step": 620
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 4.018529631194369e-05,
+      "loss": 1.4269,
+      "step": 630
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 3.989431127770635e-05,
+      "loss": 1.4133,
+      "step": 640
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 3.960016385677513e-05,
+      "loss": 1.4173,
+      "step": 650
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 3.9302916503054246e-05,
+      "loss": 1.4007,
+      "step": 660
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 3.9002632328631164e-05,
+      "loss": 1.4117,
+      "step": 670
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 3.8699375090376534e-05,
+      "loss": 1.4247,
+      "step": 680
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 3.8393209176407223e-05,
+      "loss": 1.4312,
+      "step": 690
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 3.8084199592415305e-05,
+      "loss": 1.4136,
+      "step": 700
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 3.777241194786591e-05,
+      "loss": 1.4132,
+      "step": 710
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 3.745791244206697e-05,
+      "loss": 1.4145,
+      "step": 720
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 3.714076785011359e-05,
+      "loss": 1.3951,
+      "step": 730
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 3.682104550871031e-05,
+      "loss": 1.4139,
+      "step": 740
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 3.649881330187401e-05,
+      "loss": 1.4078,
+      "step": 750
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 3.617413964652067e-05,
+      "loss": 1.4048,
+      "step": 760
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 3.5847093477938956e-05,
+      "loss": 1.4115,
+      "step": 770
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 3.551774423515378e-05,
+      "loss": 1.4018,
+      "step": 780
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 3.518616184618288e-05,
+      "loss": 1.4091,
+      "step": 790
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 3.4852416713189526e-05,
+      "loss": 1.4135,
+      "step": 800
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 3.4516579697534705e-05,
+      "loss": 1.3977,
+      "step": 810
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 3.417872210473162e-05,
+      "loss": 1.4123,
+      "step": 820
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 3.3838915669306034e-05,
+      "loss": 1.4096,
+      "step": 830
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 3.349723253956542e-05,
+      "loss": 1.4031,
+      "step": 840
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 3.315374526228036e-05,
+      "loss": 1.4104,
+      "step": 850
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 3.2808526767281225e-05,
+      "loss": 1.3972,
+      "step": 860
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 3.246165035197364e-05,
+      "loss": 1.3987,
+      "step": 870
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 3.211318966577581e-05,
+      "loss": 1.4197,
+      "step": 880
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 3.176321869448116e-05,
+      "loss": 1.4169,
+      "step": 890
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 3.1411811744549536e-05,
+      "loss": 1.4055,
+      "step": 900
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 3.105904342733032e-05,
+      "loss": 1.402,
+      "step": 910
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 3.070498864322081e-05,
+      "loss": 1.4148,
+      "step": 920
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 3.034972256576328e-05,
+      "loss": 1.4007,
+      "step": 930
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 2.999332062568395e-05,
+      "loss": 1.3926,
+      "step": 940
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 2.9635858494877384e-05,
+      "loss": 1.4099,
+      "step": 950
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 2.9277412070339782e-05,
+      "loss": 1.401,
+      "step": 960
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 2.891805745805429e-05,
+      "loss": 1.3995,
+      "step": 970
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 2.8557870956832132e-05,
+      "loss": 1.4108,
+      "step": 980
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 2.8196929042112652e-05,
+      "loss": 1.4006,
+      "step": 990
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 2.783530834972594e-05,
+      "loss": 1.4045,
+      "step": 1000
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 2.7473085659621377e-05,
+      "loss": 1.3907,
+      "step": 1010
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 2.711033787956555e-05,
+      "loss": 1.3872,
+      "step": 1020
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 2.6747142028813105e-05,
+      "loss": 1.405,
+      "step": 1030
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 2.638357522175383e-05,
+      "loss": 1.3952,
+      "step": 1040
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 2.6019714651539646e-05,
+      "loss": 1.3901,
+      "step": 1050
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 2.565563757369475e-05,
+      "loss": 1.4027,
+      "step": 1060
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 2.529142128971268e-05,
+      "loss": 1.3909,
+      "step": 1070
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.492714313064342e-05,
+      "loss": 1.4039,
+      "step": 1080
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.45628804406744e-05,
+      "loss": 1.3964,
+      "step": 1090
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.419871056070862e-05,
+      "loss": 1.4076,
+      "step": 1100
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 2.3834710811943514e-05,
+      "loss": 1.3891,
+      "step": 1110
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 2.3470958479453938e-05,
+      "loss": 1.4004,
+      "step": 1120
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 2.3107530795782877e-05,
+      "loss": 1.3879,
+      "step": 1130
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 2.2744504924543313e-05,
+      "loss": 1.3879,
+      "step": 1140
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 2.23819579440347e-05,
+      "loss": 1.3851,
+      "step": 1150
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 2.2019966830877545e-05,
+      "loss": 1.3846,
+      "step": 1160
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 2.1658608443669635e-05,
+      "loss": 1.3936,
+      "step": 1170
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 2.1297959506667224e-05,
+      "loss": 1.3841,
+      "step": 1180
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 2.0938096593494855e-05,
+      "loss": 1.3969,
+      "step": 1190
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 2.057909611088709e-05,
+      "loss": 1.3953,
+      "step": 1200
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 2.02210342824657e-05,
+      "loss": 1.3951,
+      "step": 1210
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.9863987132555706e-05,
+      "loss": 1.4012,
+      "step": 1220
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.9508030470043806e-05,
+      "loss": 1.3998,
+      "step": 1230
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 1.915323987228247e-05,
+      "loss": 1.3933,
+      "step": 1240
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.8799690669043212e-05,
+      "loss": 1.3977,
+      "step": 1250
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 1.8447457926522454e-05,
+      "loss": 1.3818,
+      "step": 1260
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.8096616431403325e-05,
+      "loss": 1.4032,
+      "step": 1270
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.7747240674976857e-05,
+      "loss": 1.3853,
+      "step": 1280
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.7399404837325796e-05,
+      "loss": 1.3988,
+      "step": 1290
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.7053182771574633e-05,
+      "loss": 1.3963,
+      "step": 1300
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.6708647988208887e-05,
+      "loss": 1.3881,
+      "step": 1310
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.6365873639467315e-05,
+      "loss": 1.3804,
+      "step": 1320
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.602493250381003e-05,
+      "loss": 1.3952,
+      "step": 1330
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.5685896970466123e-05,
+      "loss": 1.4015,
+      "step": 1340
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 1.534883902406375e-05,
+      "loss": 1.3797,
+      "step": 1350
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.5013830229346326e-05,
+      "loss": 1.3931,
+      "step": 1360
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.4680941715977722e-05,
+      "loss": 1.3946,
+      "step": 1370
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 1.4350244163439892e-05,
+      "loss": 1.3851,
+      "step": 1380
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.4021807786026108e-05,
+      "loss": 1.3891,
+      "step": 1390
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.3695702317932862e-05,
+      "loss": 1.3958,
+      "step": 1400
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.337199699845387e-05,
+      "loss": 1.3909,
+      "step": 1410
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.3050760557279023e-05,
+      "loss": 1.3785,
+      "step": 1420
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 1.2732061199901562e-05,
+      "loss": 1.3882,
+      "step": 1430
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.2415966593136547e-05,
+      "loss": 1.3872,
+      "step": 1440
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.2102543850753808e-05,
+      "loss": 1.4032,
+      "step": 1450
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.1791859519228138e-05,
+      "loss": 1.3928,
+      "step": 1460
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 1.148397956361007e-05,
+      "loss": 1.3785,
+      "step": 1470
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 1.1178969353520018e-05,
+      "loss": 1.3797,
+      "step": 1480
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.087689364926897e-05,
+      "loss": 1.3966,
+      "step": 1490
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.0577816588108378e-05,
+      "loss": 1.3849,
+      "step": 1500
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 1.0281801670612449e-05,
+      "loss": 1.3732,
+      "step": 1510
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 9.988911747195603e-06,
+      "loss": 1.4018,
+      "step": 1520
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 9.699209004767953e-06,
+      "loss": 1.3902,
+      "step": 1530
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 9.412754953531663e-06,
+      "loss": 1.3991,
+      "step": 1540
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 9.129610413921005e-06,
+      "loss": 1.3824,
+      "step": 1550
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 8.849835503688847e-06,
+      "loss": 1.3904,
+      "step": 1560
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 8.57348962514234e-06,
+      "loss": 1.3937,
+      "step": 1570
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 8.30063145253053e-06,
+      "loss": 1.39,
+      "step": 1580
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 8.031318919586523e-06,
+      "loss": 1.3773,
+      "step": 1590
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 7.7656092072269e-06,
+      "loss": 1.3875,
+      "step": 1600
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 7.503558731410959e-06,
+      "loss": 1.3859,
+      "step": 1610
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 7.245223131162376e-06,
+      "loss": 1.3716,
+      "step": 1620
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 6.9906572567558285e-06,
+      "loss": 1.3846,
+      "step": 1630
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 6.739915158071106e-06,
+      "loss": 1.3741,
+      "step": 1640
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 6.493050073117116e-06,
+      "loss": 1.3827,
+      "step": 1650
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 6.250114416728298e-06,
+      "loss": 1.3951,
+      "step": 1660
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 6.011159769435823e-06,
+      "loss": 1.3875,
+      "step": 1670
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 5.776236866515947e-06,
+      "loss": 1.3783,
+      "step": 1680
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 5.545395587217763e-06,
+      "loss": 1.3888,
+      "step": 1690
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 5.318684944172752e-06,
+      "loss": 1.4006,
+      "step": 1700
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 5.096153072988366e-06,
+      "loss": 1.4041,
+      "step": 1710
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 4.8778472220277874e-06,
+      "loss": 1.3865,
+      "step": 1720
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 4.6638137423780395e-06,
+      "loss": 1.3789,
+      "step": 1730
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 4.454098078008667e-06,
+      "loss": 1.3861,
+      "step": 1740
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 4.248744756122986e-06,
+      "loss": 1.3803,
+      "step": 1750
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 4.047797377703985e-06,
+      "loss": 1.3695,
+      "step": 1760
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 3.851298608256892e-06,
+      "loss": 1.3937,
+      "step": 1770
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 3.6592901687503566e-06,
+      "loss": 1.3865,
+      "step": 1780
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 3.471812826758178e-06,
+      "loss": 1.3799,
+      "step": 1790
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 3.288906387803464e-06,
+      "loss": 1.3911,
+      "step": 1800
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 3.1106096869070483e-06,
+      "loss": 1.3801,
+      "step": 1810
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 2.9369605803419715e-06,
+      "loss": 1.382,
+      "step": 1820
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 2.767995937595766e-06,
+      "loss": 1.3898,
+      "step": 1830
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 2.6037516335422728e-06,
+      "loss": 1.3752,
+      "step": 1840
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 2.4442625408246074e-06,
+      "loss": 1.3855,
+      "step": 1850
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 2.289562522450947e-06,
+      "loss": 1.3665,
+      "step": 1860
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 2.1396844246046903e-06,
+      "loss": 1.3747,
+      "step": 1870
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 1.9946600696704592e-06,
+      "loss": 1.3861,
+      "step": 1880
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 1.8545202494775509e-06,
+      "loss": 1.3844,
+      "step": 1890
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 1.7192947187621434e-06,
+      "loss": 1.3796,
+      "step": 1900
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 1.5890121888497366e-06,
+      "loss": 1.3963,
+      "step": 1910
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 1.463700321559075e-06,
+      "loss": 1.385,
+      "step": 1920
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 1.3433857233289714e-06,
+      "loss": 1.3811,
+      "step": 1930
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 1.2280939395691859e-06,
+      "loss": 1.384,
+      "step": 1940
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 1.1178494492365465e-06,
+      "loss": 1.3821,
+      "step": 1950
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 1.0126756596375686e-06,
+      "loss": 1.3978,
+      "step": 1960
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 9.125949014585383e-07,
+      "loss": 1.3724,
+      "step": 1970
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 8.176284240242638e-07,
+      "loss": 1.3851,
+      "step": 1980
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 7.277963907863478e-07,
+      "loss": 1.3793,
+      "step": 1990
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 6.431178750420513e-07,
+      "loss": 1.3946,
+      "step": 2000
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 5.636108558846359e-07,
+      "loss": 1.3794,
+      "step": 2010
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 4.892922143859918e-07,
+      "loss": 1.3886,
+      "step": 2020
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 4.201777300124249e-07,
+      "loss": 1.3794,
+      "step": 2030
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 3.562820772743392e-07,
+      "loss": 1.3828,
+      "step": 2040
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 2.9761882261050143e-07,
+      "loss": 1.374,
+      "step": 2050
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 2.4420042150761215e-07,
+      "loss": 1.3804,
+      "step": 2060
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 1.9603821585572058e-07,
+      "loss": 1.3772,
+      "step": 2070
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 1.531424315400931e-07,
+      "loss": 1.3913,
+      "step": 2080
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 1.1552217627004425e-07,
+      "loss": 1.3859,
+      "step": 2090
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 8.318543764516961e-08,
+      "loss": 1.4035,
+      "step": 2100
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 5.613908145939428e-08,
+      "loss": 1.4047,
+      "step": 2110
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 3.438885024322769e-08,
+      "loss": 1.3886,
+      "step": 2120
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 1.7939362044494534e-08,
+      "loss": 1.3874,
+      "step": 2130
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 6.794109447824659e-09,
+      "loss": 1.3852,
+      "step": 2140
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 9.554588330934478e-10,
+      "loss": 1.3859,
+      "step": 2150
+    },
+    {
+      "epoch": 2.0,
+      "step": 2156,
+      "total_flos": 4.637442392914395e+18,
+      "train_loss": 1.4111853902105498,
+      "train_runtime": 51425.1791,
+      "train_samples_per_second": 2.684,
+      "train_steps_per_second": 0.042
+    }
+  ],
+  "max_steps": 2156,
+  "num_train_epochs": 2,
+  "total_flos": 4.637442392914395e+18,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:918a94b2b648985fdb03a840052eff79b0035343fd87ee2a0dc751e904fad0c4
+size 4091