add

README.md

@@ -0,0 +1,21 @@
+---
+library_name: peft
+---
+## Training procedure
+
+
+The following `bitsandbytes` quantization config was used during training:
+- quant_method: bitsandbytes
+- 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 +1,20 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": "./Qwen-14B-Chat",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "lora_alpha": 32.0,
+  "lora_dropout": 0.1,
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 8,
+  "revision": null,
+  "target_modules": [
+    "c_attn"
+  ],
+  "task_type": "CAUSAL_LM"
+}

adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8f8bc51e048df5ba34d3ae72bebbaa86dd4da5a06de34a514834d9e5d6c6c10c
+size 26242657

all_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 2.0,
+    "train_loss": 1.2461530792912217,
+    "train_runtime": 30560.3748,
+    "train_samples_per_second": 4.517,
+    "train_steps_per_second": 0.071
+}

checkpoint-1000/README.md

@@ -0,0 +1,21 @@
+---
+library_name: peft
+---
+## Training procedure
+
+
+The following `bitsandbytes` quantization config was used during training:
+- quant_method: bitsandbytes
+- 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 +1,20 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": "./Qwen-14B-Chat",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "lora_alpha": 32.0,
+  "lora_dropout": 0.1,
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 8,
+  "revision": null,
+  "target_modules": [
+    "c_attn"
+  ],
+  "task_type": "CAUSAL_LM"
+}

checkpoint-1000/adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:829bd89f4173976589d1b69f5c878ab86c2d7b9e825d7f36acc3767fd7fbed3d
+size 26242657

checkpoint-1000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2e1f2e79f1fd8938b95d2f5ac67a77bdef04659b59547d26bfaa6e5e0be354d
+size 52496005

checkpoint-1000/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:28242f711bbe2bade0d1b941200359466639a388ffaa5ae87f80e2b210be5ef4
+size 18679

checkpoint-1000/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:53a4add8376116454f4f747521df1d370a365bad8ac691065ced3f2dc972dd8a
+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 +1,7 @@
+{
+  "additional_special_tokens": [
+    "<|im_end|>"
+  ],
+  "eos_token": "<|endoftext|>",
+  "pad_token": "<|endoftext|>"
+}

checkpoint-1000/tokenization_qwen.py

@@ -0,0 +1,246 @@
+# Copyright (c) Alibaba Cloud.
+#
+# This source code is licensed under the license found in the
+# LICENSE file in the root directory of this source tree.
+
+"""Tokenization classes for QWen."""
+
+import base64
+import logging
+import os
+import unicodedata
+from typing import Collection, Dict, List, Set, Tuple, Union
+
+import tiktoken
+from transformers import PreTrainedTokenizer, AddedToken
+
+logger = logging.getLogger(__name__)
+
+
+VOCAB_FILES_NAMES = {"vocab_file": "qwen.tiktoken"}
+
+PAT_STR = r"""(?i:'s|'t|'re|'ve|'m|'ll|'d)|[^\r\n\p{L}\p{N}]?\p{L}+|\p{N}| ?[^\s\p{L}\p{N}]+[\r\n]*|\s*[\r\n]+|\s+(?!\S)|\s+"""
+ENDOFTEXT = "<|endoftext|>"
+IMSTART = "<|im_start|>"
+IMEND = "<|im_end|>"
+# as the default behavior is changed to allow special tokens in
+# regular texts, the surface forms of special tokens need to be
+# as different as possible to minimize the impact
+EXTRAS = tuple((f"<|extra_{i}|>" for i in range(205)))
+SPECIAL_TOKENS = (
+    ENDOFTEXT,
+    IMSTART,
+    IMEND,
+) + EXTRAS
+
+
+def _load_tiktoken_bpe(tiktoken_bpe_file: str) -> Dict[bytes, int]:
+    with open(tiktoken_bpe_file, "rb") as f:
+        contents = f.read()
+    return {
+        base64.b64decode(token): int(rank)
+        for token, rank in (line.split() for line in contents.splitlines() if line)
+    }
+
+class QWenTokenizer(PreTrainedTokenizer):
+    """QWen tokenizer."""
+
+    vocab_files_names = VOCAB_FILES_NAMES
+
+    def __init__(
+        self,
+        vocab_file,
+        errors="replace",
+        **kwargs,
+    ):
+        super().__init__(**kwargs)
+
+        self.errors = errors  # how to handle errors in decoding
+
+        self.mergeable_ranks = _load_tiktoken_bpe(vocab_file)  # type: dict[bytes, int]
+        self.special_tokens = {
+            token: index
+            for index, token in enumerate(
+                SPECIAL_TOKENS, start=len(self.mergeable_ranks)
+            )
+        }
+
+        enc = tiktoken.Encoding(
+            "Qwen",
+            pat_str=PAT_STR,
+            mergeable_ranks=self.mergeable_ranks,
+            special_tokens=self.special_tokens,
+        )
+        assert (
+            len(self.mergeable_ranks) + len(self.special_tokens) == enc.n_vocab
+        ), f"{len(self.mergeable_ranks) + len(self.special_tokens)} != {enc.n_vocab} in encoding"
+
+        self.decoder = {
+            v: k for k, v in self.mergeable_ranks.items()
+        }  # type: dict[int, bytes|str]
+        self.decoder.update({v: k for k, v in self.special_tokens.items()})
+
+        self.tokenizer = enc  # type: tiktoken.Encoding
+
+        self.eod_id = self.tokenizer.eot_token
+        self.im_start_id = self.special_tokens[IMSTART]
+        self.im_end_id = self.special_tokens[IMEND]
+
+    def __getstate__(self):
+        # for pickle lovers
+        state = self.__dict__.copy()
+        del state['tokenizer']
+        return state
+
+    def __setstate__(self, state):
+        # tokenizer is not python native; don't pass it; rebuild it
+        self.__dict__.update(state)
+        enc = tiktoken.Encoding(
+            "Qwen",
+            pat_str=PAT_STR,
+            mergeable_ranks=self.mergeable_ranks,
+            special_tokens=self.special_tokens,
+        )
+        self.tokenizer = enc
+
+
+    def __len__(self) -> int:
+        return self.tokenizer.n_vocab
+
+    def get_vocab(self) -> Dict[bytes, int]:
+        return self.mergeable_ranks
+
+    def convert_tokens_to_ids(
+        self, tokens: Union[bytes, str, List[Union[bytes, str]]]
+    ) -> List[int]:
+        ids = []
+        if isinstance(tokens, (str, bytes)):
+            if tokens in self.special_tokens:
+                return self.special_tokens[tokens]
+            else:
+                return self.mergeable_ranks.get(tokens)
+        for token in tokens:
+            if token in self.special_tokens:
+                ids.append(self.special_tokens[token])
+            else:
+                ids.append(self.mergeable_ranks.get(token))
+        return ids
+
+    def _add_tokens(self, new_tokens: Union[List[str], List[AddedToken]], special_tokens: bool = False) -> int:
+        if not special_tokens and new_tokens:
+            raise ValueError('Adding regular tokens is not supported')
+        for token in new_tokens:
+            surface_form = token.content if isinstance(token, AddedToken) else token
+            if surface_form not in SPECIAL_TOKENS:
+                raise ValueError('Adding unknown special tokens is not supported')
+        return 0
+
+    def save_vocabulary(self, save_directory: str, **kwargs) -> Tuple[str]:
+        """
+        Save only the vocabulary of the tokenizer (vocabulary).
+
+        Returns:
+            `Tuple(str)`: Paths to the files saved.
+        """
+        file_path = os.path.join(save_directory, "qwen.tiktoken")
+        with open(file_path, "w", encoding="utf8") as w:
+            for k, v in self.mergeable_ranks.items():
+                line = base64.b64encode(k).decode("utf8") + " " + str(v) + "\n"
+                w.write(line)
+        return (file_path,)
+
+    def tokenize(
+        self,
+        text: str,
+        allowed_special: Union[Set, str] = "all",
+        disallowed_special: Union[Collection, str] = (),
+        **kwargs,
+    ) -> List[Union[bytes, str]]:
+        """
+        Converts a string in a sequence of tokens.
+
+        Args:
+            text (`str`):
+                The sequence to be encoded.
+            allowed_special (`Literal["all"]` or `set`):
+                The surface forms of the tokens to be encoded as special tokens in regular texts.
+                Default to "all".
+            disallowed_special (`Literal["all"]` or `Collection`):
+                The surface forms of the tokens that should not be in regular texts and trigger errors.
+                Default to an empty tuple.
+
+            kwargs (additional keyword arguments, *optional*):
+                Will be passed to the underlying model specific encode method.
+
+        Returns:
+            `List[bytes|str]`: The list of tokens.
+        """
+        tokens = []
+        text = unicodedata.normalize("NFC", text)
+
+        # this implementation takes a detour: text -> token id -> token surface forms
+        for t in self.tokenizer.encode(
+            text, allowed_special=allowed_special, disallowed_special=disallowed_special
+        ):
+            tokens.append(self.decoder[t])
+        return tokens
+
+    def convert_tokens_to_string(self, tokens: List[Union[bytes, str]]) -> str:
+        """
+        Converts a sequence of tokens in a single string.
+        """
+        text = ""
+        temp = b""
+        for t in tokens:
+            if isinstance(t, str):
+                if temp:
+                    text += temp.decode("utf-8", errors=self.errors)
+                    temp = b""
+                text += t
+            elif isinstance(t, bytes):
+                temp += t
+            else:
+                raise TypeError("token should only be of type types or str")
+        if temp:
+            text += temp.decode("utf-8", errors=self.errors)
+        return text
+
+    @property
+    def vocab_size(self):
+        return self.tokenizer.n_vocab
+
+    def _convert_id_to_token(self, index: int) -> Union[bytes, str]:
+        """Converts an id to a token, special tokens included"""
+        if index in self.decoder:
+            return self.decoder[index]
+        raise ValueError("unknown ids")
+
+    def _convert_token_to_id(self, token: Union[bytes, str]) -> int:
+        """Converts a token to an id using the vocab, special tokens included"""
+        if token in self.special_tokens:
+            return self.special_tokens[token]
+        if token in self.mergeable_ranks:
+            return self.mergeable_ranks[token]
+        raise ValueError("unknown token")
+
+    def _tokenize(self, text: str, **kwargs):
+        """
+        Converts a string in a sequence of tokens (string), using the tokenizer. Split in words for word-based
+        vocabulary or sub-words for sub-word-based vocabularies (BPE/SentencePieces/WordPieces).
+
+        Do NOT take care of added tokens.
+        """
+        raise NotImplementedError
+
+    def _decode(
+        self,
+        token_ids: Union[int, List[int]],
+        skip_special_tokens: bool = False,
+        errors: str = None,
+        **kwargs,
+    ) -> str:
+        if isinstance(token_ids, int):
+            token_ids = [token_ids]
+        if skip_special_tokens:
+            token_ids = [i for i in token_ids if i < self.eod_id]
+        return self.tokenizer.decode(token_ids, errors=errors or self.errors)

checkpoint-1000/tokenizer_config.json

@@ -0,0 +1,12 @@
+{
+  "auto_map": {
+    "AutoTokenizer": [
+      "tokenization_qwen.QWenTokenizer",
+      null
+    ]
+  },
+  "clean_up_tokenization_spaces": true,
+  "model_max_length": 8192,
+  "padding_side": "right",
+  "tokenizer_class": "QWenTokenizer"
+}

checkpoint-1000/trainer_state.json

@@ -0,0 +1,619 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.9272137227630969,
+  "eval_steps": 500,
+  "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": 1.6199,
+      "step": 10
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.998938447446803e-05,
+      "loss": 1.5062,
+      "step": 20
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 4.997611718058365e-05,
+      "loss": 1.4138,
+      "step": 30
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.9957546913022665e-05,
+      "loss": 1.369,
+      "step": 40
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 4.993367761465736e-05,
+      "loss": 1.3408,
+      "step": 50
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.9904514353459654e-05,
+      "loss": 1.3321,
+      "step": 60
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.9870063321425105e-05,
+      "loss": 1.3251,
+      "step": 70
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 4.983033183325818e-05,
+      "loss": 1.3228,
+      "step": 80
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 4.97853283248192e-05,
+      "loss": 1.3111,
+      "step": 90
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 4.973506235133323e-05,
+      "loss": 1.3013,
+      "step": 100
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 4.967954458536126e-05,
+      "loss": 1.3004,
+      "step": 110
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 4.9618786814534226e-05,
+      "loss": 1.2959,
+      "step": 120
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 4.955280193905022e-05,
+      "loss": 1.2969,
+      "step": 130
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.948160396893553e-05,
+      "loss": 1.2879,
+      "step": 140
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 4.9405208021069946e-05,
+      "loss": 1.277,
+      "step": 150
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 4.9323630315977156e-05,
+      "loss": 1.283,
+      "step": 160
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.9236888174380784e-05,
+      "loss": 1.288,
+      "step": 170
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 4.91450000135268e-05,
+      "loss": 1.287,
+      "step": 180
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.9047985343273154e-05,
+      "loss": 1.2726,
+      "step": 190
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.894586476194739e-05,
+      "loss": 1.2808,
+      "step": 200
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.883865995197319e-05,
+      "loss": 1.2657,
+      "step": 210
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.8726393675266716e-05,
+      "loss": 1.275,
+      "step": 220
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 4.860908976840376e-05,
+      "loss": 1.2667,
+      "step": 230
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 4.848677313755872e-05,
+      "loss": 1.2715,
+      "step": 240
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 4.835946975321647e-05,
+      "loss": 1.273,
+      "step": 250
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 4.822720664465827e-05,
+      "loss": 1.2605,
+      "step": 260
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 4.809001189422287e-05,
+      "loss": 1.2766,
+      "step": 270
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 4.794791463134399e-05,
+      "loss": 1.262,
+      "step": 280
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 4.780094502636552e-05,
+      "loss": 1.255,
+      "step": 290
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 4.764913428413572e-05,
+      "loss": 1.2652,
+      "step": 300
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 4.7492514637381727e-05,
+      "loss": 1.2668,
+      "step": 310
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 4.733111933986583e-05,
+      "loss": 1.2621,
+      "step": 320
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 4.716498265932501e-05,
+      "loss": 1.257,
+      "step": 330
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 4.699413987019512e-05,
+      "loss": 1.2789,
+      "step": 340
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 4.681862724612141e-05,
+      "loss": 1.2634,
+      "step": 350
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 4.663848205225674e-05,
+      "loss": 1.2594,
+      "step": 360
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 4.645374253734949e-05,
+      "loss": 1.26,
+      "step": 370
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 4.626444792562244e-05,
+      "loss": 1.2514,
+      "step": 380
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 4.607063840844463e-05,
+      "loss": 1.2506,
+      "step": 390
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 4.587235513579791e-05,
+      "loss": 1.2648,
+      "step": 400
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 4.5669640207539786e-05,
+      "loss": 1.2511,
+      "step": 410
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 4.546253666446484e-05,
+      "loss": 1.2594,
+      "step": 420
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 4.525108847916614e-05,
+      "loss": 1.2608,
+      "step": 430
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 4.503534054669892e-05,
+      "loss": 1.2597,
+      "step": 440
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 4.481533867504841e-05,
+      "loss": 1.2609,
+      "step": 450
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 4.4591129575403765e-05,
+      "loss": 1.2505,
+      "step": 460
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 4.43627608522403e-05,
+      "loss": 1.2481,
+      "step": 470
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 4.4130280993211974e-05,
+      "loss": 1.2612,
+      "step": 480
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 4.389373935885646e-05,
+      "loss": 1.2504,
+      "step": 490
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 4.365318617211479e-05,
+      "loss": 1.2518,
+      "step": 500
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 4.340867250766794e-05,
+      "loss": 1.2458,
+      "step": 510
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 4.316025028109258e-05,
+      "loss": 1.2345,
+      "step": 520
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 4.2907972237838225e-05,
+      "loss": 1.2521,
+      "step": 530
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 4.2651891942028274e-05,
+      "loss": 1.2528,
+      "step": 540
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 4.239206376508717e-05,
+      "loss": 1.2462,
+      "step": 550
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 4.212854287419611e-05,
+      "loss": 1.2401,
+      "step": 560
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 4.1861385220579934e-05,
+      "loss": 1.2496,
+      "step": 570
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 4.1590647527627404e-05,
+      "loss": 1.2522,
+      "step": 580
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 4.131638727884762e-05,
+      "loss": 1.2377,
+      "step": 590
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 4.103866270566498e-05,
+      "loss": 1.2467,
+      "step": 600
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 4.075753277505544e-05,
+      "loss": 1.2421,
+      "step": 610
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 4.0473057177026484e-05,
+      "loss": 1.2455,
+      "step": 620
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 4.018529631194369e-05,
+      "loss": 1.2294,
+      "step": 630
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 3.989431127770635e-05,
+      "loss": 1.2509,
+      "step": 640
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 3.960016385677513e-05,
+      "loss": 1.2354,
+      "step": 650
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 3.9302916503054246e-05,
+      "loss": 1.2318,
+      "step": 660
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 3.9002632328631164e-05,
+      "loss": 1.2376,
+      "step": 670
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 3.8699375090376534e-05,
+      "loss": 1.2412,
+      "step": 680
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 3.8393209176407223e-05,
+      "loss": 1.2479,
+      "step": 690
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 3.8084199592415305e-05,
+      "loss": 1.2431,
+      "step": 700
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 3.777241194786591e-05,
+      "loss": 1.245,
+      "step": 710
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 3.745791244206697e-05,
+      "loss": 1.2393,
+      "step": 720
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 3.714076785011359e-05,
+      "loss": 1.2473,
+      "step": 730
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 3.682104550871031e-05,
+      "loss": 1.2552,
+      "step": 740
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 3.649881330187401e-05,
+      "loss": 1.2356,
+      "step": 750
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 3.617413964652067e-05,
+      "loss": 1.2442,
+      "step": 760
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 3.5847093477938956e-05,
+      "loss": 1.2605,
+      "step": 770
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 3.551774423515378e-05,
+      "loss": 1.2364,
+      "step": 780
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 3.518616184618288e-05,
+      "loss": 1.2482,
+      "step": 790
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 3.4852416713189526e-05,
+      "loss": 1.2327,
+      "step": 800
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 3.4516579697534705e-05,
+      "loss": 1.2548,
+      "step": 810
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 3.417872210473162e-05,
+      "loss": 1.2469,
+      "step": 820
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 3.3838915669306034e-05,
+      "loss": 1.2424,
+      "step": 830
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 3.349723253956542e-05,
+      "loss": 1.2438,
+      "step": 840
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 3.315374526228036e-05,
+      "loss": 1.243,
+      "step": 850
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 3.2808526767281225e-05,
+      "loss": 1.2416,
+      "step": 860
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 3.246165035197364e-05,
+      "loss": 1.2399,
+      "step": 870
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 3.211318966577581e-05,
+      "loss": 1.239,
+      "step": 880
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 3.176321869448116e-05,
+      "loss": 1.2498,
+      "step": 890
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 3.1411811744549536e-05,
+      "loss": 1.2438,
+      "step": 900
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 3.105904342733032e-05,
+      "loss": 1.233,
+      "step": 910
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 3.070498864322081e-05,
+      "loss": 1.2315,
+      "step": 920
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 3.034972256576328e-05,
+      "loss": 1.232,
+      "step": 930
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 2.999332062568395e-05,
+      "loss": 1.2467,
+      "step": 940
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 2.9635858494877384e-05,
+      "loss": 1.2483,
+      "step": 950
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 2.9277412070339782e-05,
+      "loss": 1.2377,
+      "step": 960
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 2.891805745805429e-05,
+      "loss": 1.2347,
+      "step": 970
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 2.8557870956832132e-05,
+      "loss": 1.2428,
+      "step": 980
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 2.8196929042112652e-05,
+      "loss": 1.244,
+      "step": 990
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 2.783530834972594e-05,
+      "loss": 1.2317,
+      "step": 1000
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 2156,
+  "num_train_epochs": 2,
+  "save_steps": 1000,
+  "total_flos": 1.5703812576473252e+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:e1ca0e4372bd1720a48da585ce6946399a06d0c984dcfcf5a2b7104827c4603f
+size 4219

checkpoint-2000/README.md

@@ -0,0 +1,21 @@
+---
+library_name: peft
+---
+## Training procedure
+
+
+The following `bitsandbytes` quantization config was used during training:
+- quant_method: bitsandbytes
+- 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 +1,20 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": "./Qwen-14B-Chat",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "lora_alpha": 32.0,
+  "lora_dropout": 0.1,
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 8,
+  "revision": null,
+  "target_modules": [
+    "c_attn"
+  ],
+  "task_type": "CAUSAL_LM"
+}

checkpoint-2000/adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fd24a9b391571bbdc9f88c1c9433cdd04de9e12bc47315124781f1fbb40b768f
+size 26242657

checkpoint-2000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:310608e62ebf23c5f12f1b89ef6037683c0938b0c464e55f1186e182461eaff9
+size 52496005

checkpoint-2000/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:402d9b1bf0a32ecfe8861b4f22edf733a41b828d3d32764e5ccad98767ebbf20
+size 18679

checkpoint-2000/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c1cc36814e5cb2cf8d7e0bc2c24d4362c7a781368af8673695b07b0575ebaa8
+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 +1,7 @@
+{
+  "additional_special_tokens": [
+    "<|im_end|>"
+  ],
+  "eos_token": "<|endoftext|>",
+  "pad_token": "<|endoftext|>"
+}

checkpoint-2000/tokenization_qwen.py

@@ -0,0 +1,246 @@
+# Copyright (c) Alibaba Cloud.
+#
+# This source code is licensed under the license found in the
+# LICENSE file in the root directory of this source tree.
+
+"""Tokenization classes for QWen."""
+
+import base64
+import logging
+import os
+import unicodedata
+from typing import Collection, Dict, List, Set, Tuple, Union
+
+import tiktoken
+from transformers import PreTrainedTokenizer, AddedToken
+
+logger = logging.getLogger(__name__)
+
+
+VOCAB_FILES_NAMES = {"vocab_file": "qwen.tiktoken"}
+
+PAT_STR = r"""(?i:'s|'t|'re|'ve|'m|'ll|'d)|[^\r\n\p{L}\p{N}]?\p{L}+|\p{N}| ?[^\s\p{L}\p{N}]+[\r\n]*|\s*[\r\n]+|\s+(?!\S)|\s+"""
+ENDOFTEXT = "<|endoftext|>"
+IMSTART = "<|im_start|>"
+IMEND = "<|im_end|>"
+# as the default behavior is changed to allow special tokens in
+# regular texts, the surface forms of special tokens need to be
+# as different as possible to minimize the impact
+EXTRAS = tuple((f"<|extra_{i}|>" for i in range(205)))
+SPECIAL_TOKENS = (
+    ENDOFTEXT,
+    IMSTART,
+    IMEND,
+) + EXTRAS
+
+
+def _load_tiktoken_bpe(tiktoken_bpe_file: str) -> Dict[bytes, int]:
+    with open(tiktoken_bpe_file, "rb") as f:
+        contents = f.read()
+    return {
+        base64.b64decode(token): int(rank)
+        for token, rank in (line.split() for line in contents.splitlines() if line)
+    }
+
+class QWenTokenizer(PreTrainedTokenizer):
+    """QWen tokenizer."""
+
+    vocab_files_names = VOCAB_FILES_NAMES
+
+    def __init__(
+        self,
+        vocab_file,
+        errors="replace",
+        **kwargs,
+    ):
+        super().__init__(**kwargs)
+
+        self.errors = errors  # how to handle errors in decoding
+
+        self.mergeable_ranks = _load_tiktoken_bpe(vocab_file)  # type: dict[bytes, int]
+        self.special_tokens = {
+            token: index
+            for index, token in enumerate(
+                SPECIAL_TOKENS, start=len(self.mergeable_ranks)
+            )
+        }
+
+        enc = tiktoken.Encoding(
+            "Qwen",
+            pat_str=PAT_STR,
+            mergeable_ranks=self.mergeable_ranks,
+            special_tokens=self.special_tokens,
+        )
+        assert (
+            len(self.mergeable_ranks) + len(self.special_tokens) == enc.n_vocab
+        ), f"{len(self.mergeable_ranks) + len(self.special_tokens)} != {enc.n_vocab} in encoding"
+
+        self.decoder = {
+            v: k for k, v in self.mergeable_ranks.items()
+        }  # type: dict[int, bytes|str]
+        self.decoder.update({v: k for k, v in self.special_tokens.items()})
+
+        self.tokenizer = enc  # type: tiktoken.Encoding
+
+        self.eod_id = self.tokenizer.eot_token
+        self.im_start_id = self.special_tokens[IMSTART]
+        self.im_end_id = self.special_tokens[IMEND]
+
+    def __getstate__(self):
+        # for pickle lovers
+        state = self.__dict__.copy()
+        del state['tokenizer']
+        return state
+
+    def __setstate__(self, state):
+        # tokenizer is not python native; don't pass it; rebuild it
+        self.__dict__.update(state)
+        enc = tiktoken.Encoding(
+            "Qwen",
+            pat_str=PAT_STR,
+            mergeable_ranks=self.mergeable_ranks,
+            special_tokens=self.special_tokens,
+        )
+        self.tokenizer = enc
+
+
+    def __len__(self) -> int:
+        return self.tokenizer.n_vocab
+
+    def get_vocab(self) -> Dict[bytes, int]:
+        return self.mergeable_ranks
+
+    def convert_tokens_to_ids(
+        self, tokens: Union[bytes, str, List[Union[bytes, str]]]
+    ) -> List[int]:
+        ids = []
+        if isinstance(tokens, (str, bytes)):
+            if tokens in self.special_tokens:
+                return self.special_tokens[tokens]
+            else:
+                return self.mergeable_ranks.get(tokens)
+        for token in tokens:
+            if token in self.special_tokens:
+                ids.append(self.special_tokens[token])
+            else:
+                ids.append(self.mergeable_ranks.get(token))
+        return ids
+
+    def _add_tokens(self, new_tokens: Union[List[str], List[AddedToken]], special_tokens: bool = False) -> int:
+        if not special_tokens and new_tokens:
+            raise ValueError('Adding regular tokens is not supported')
+        for token in new_tokens:
+            surface_form = token.content if isinstance(token, AddedToken) else token
+            if surface_form not in SPECIAL_TOKENS:
+                raise ValueError('Adding unknown special tokens is not supported')
+        return 0
+
+    def save_vocabulary(self, save_directory: str, **kwargs) -> Tuple[str]:
+        """
+        Save only the vocabulary of the tokenizer (vocabulary).
+
+        Returns:
+            `Tuple(str)`: Paths to the files saved.
+        """
+        file_path = os.path.join(save_directory, "qwen.tiktoken")
+        with open(file_path, "w", encoding="utf8") as w:
+            for k, v in self.mergeable_ranks.items():
+                line = base64.b64encode(k).decode("utf8") + " " + str(v) + "\n"
+                w.write(line)
+        return (file_path,)
+
+    def tokenize(
+        self,
+        text: str,
+        allowed_special: Union[Set, str] = "all",
+        disallowed_special: Union[Collection, str] = (),
+        **kwargs,
+    ) -> List[Union[bytes, str]]:
+        """
+        Converts a string in a sequence of tokens.
+
+        Args:
+            text (`str`):
+                The sequence to be encoded.
+            allowed_special (`Literal["all"]` or `set`):
+                The surface forms of the tokens to be encoded as special tokens in regular texts.
+                Default to "all".
+            disallowed_special (`Literal["all"]` or `Collection`):
+                The surface forms of the tokens that should not be in regular texts and trigger errors.
+                Default to an empty tuple.
+
+            kwargs (additional keyword arguments, *optional*):
+                Will be passed to the underlying model specific encode method.
+
+        Returns:
+            `List[bytes|str]`: The list of tokens.
+        """
+        tokens = []
+        text = unicodedata.normalize("NFC", text)
+
+        # this implementation takes a detour: text -> token id -> token surface forms
+        for t in self.tokenizer.encode(
+            text, allowed_special=allowed_special, disallowed_special=disallowed_special
+        ):
+            tokens.append(self.decoder[t])
+        return tokens
+
+    def convert_tokens_to_string(self, tokens: List[Union[bytes, str]]) -> str:
+        """
+        Converts a sequence of tokens in a single string.
+        """
+        text = ""
+        temp = b""
+        for t in tokens:
+            if isinstance(t, str):
+                if temp:
+                    text += temp.decode("utf-8", errors=self.errors)
+                    temp = b""
+                text += t
+            elif isinstance(t, bytes):
+                temp += t
+            else:
+                raise TypeError("token should only be of type types or str")
+        if temp:
+            text += temp.decode("utf-8", errors=self.errors)
+        return text
+
+    @property
+    def vocab_size(self):
+        return self.tokenizer.n_vocab
+
+    def _convert_id_to_token(self, index: int) -> Union[bytes, str]:
+        """Converts an id to a token, special tokens included"""
+        if index in self.decoder:
+            return self.decoder[index]
+        raise ValueError("unknown ids")
+
+    def _convert_token_to_id(self, token: Union[bytes, str]) -> int:
+        """Converts a token to an id using the vocab, special tokens included"""
+        if token in self.special_tokens:
+            return self.special_tokens[token]
+        if token in self.mergeable_ranks:
+            return self.mergeable_ranks[token]
+        raise ValueError("unknown token")
+
+    def _tokenize(self, text: str, **kwargs):
+        """
+        Converts a string in a sequence of tokens (string), using the tokenizer. Split in words for word-based
+        vocabulary or sub-words for sub-word-based vocabularies (BPE/SentencePieces/WordPieces).
+
+        Do NOT take care of added tokens.
+        """
+        raise NotImplementedError
+
+    def _decode(
+        self,
+        token_ids: Union[int, List[int]],
+        skip_special_tokens: bool = False,
+        errors: str = None,
+        **kwargs,
+    ) -> str:
+        if isinstance(token_ids, int):
+            token_ids = [token_ids]
+        if skip_special_tokens:
+            token_ids = [i for i in token_ids if i < self.eod_id]
+        return self.tokenizer.decode(token_ids, errors=errors or self.errors)

checkpoint-2000/tokenizer_config.json

@@ -0,0 +1,12 @@
+{
+  "auto_map": {
+    "AutoTokenizer": [
+      "tokenization_qwen.QWenTokenizer",
+      null
+    ]
+  },
+  "clean_up_tokenization_spaces": true,
+  "model_max_length": 8192,
+  "padding_side": "right",
+  "tokenizer_class": "QWenTokenizer"
+}

checkpoint-2000/trainer_state.json

@@ -0,0 +1,1219 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.8544274455261938,
+  "eval_steps": 500,
+  "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": 1.6199,
+      "step": 10
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.998938447446803e-05,
+      "loss": 1.5062,
+      "step": 20
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 4.997611718058365e-05,
+      "loss": 1.4138,
+      "step": 30
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.9957546913022665e-05,
+      "loss": 1.369,
+      "step": 40
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 4.993367761465736e-05,
+      "loss": 1.3408,
+      "step": 50
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.9904514353459654e-05,
+      "loss": 1.3321,
+      "step": 60
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.9870063321425105e-05,
+      "loss": 1.3251,
+      "step": 70
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 4.983033183325818e-05,
+      "loss": 1.3228,
+      "step": 80
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 4.97853283248192e-05,
+      "loss": 1.3111,
+      "step": 90
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 4.973506235133323e-05,
+      "loss": 1.3013,
+      "step": 100
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 4.967954458536126e-05,
+      "loss": 1.3004,
+      "step": 110
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 4.9618786814534226e-05,
+      "loss": 1.2959,
+      "step": 120
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 4.955280193905022e-05,
+      "loss": 1.2969,
+      "step": 130
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.948160396893553e-05,
+      "loss": 1.2879,
+      "step": 140
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 4.9405208021069946e-05,
+      "loss": 1.277,
+      "step": 150
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 4.9323630315977156e-05,
+      "loss": 1.283,
+      "step": 160
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.9236888174380784e-05,
+      "loss": 1.288,
+      "step": 170
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 4.91450000135268e-05,
+      "loss": 1.287,
+      "step": 180
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.9047985343273154e-05,
+      "loss": 1.2726,
+      "step": 190
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.894586476194739e-05,
+      "loss": 1.2808,
+      "step": 200
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.883865995197319e-05,
+      "loss": 1.2657,
+      "step": 210
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.8726393675266716e-05,
+      "loss": 1.275,
+      "step": 220
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 4.860908976840376e-05,
+      "loss": 1.2667,
+      "step": 230
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 4.848677313755872e-05,
+      "loss": 1.2715,
+      "step": 240
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 4.835946975321647e-05,
+      "loss": 1.273,
+      "step": 250
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 4.822720664465827e-05,
+      "loss": 1.2605,
+      "step": 260
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 4.809001189422287e-05,
+      "loss": 1.2766,
+      "step": 270
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 4.794791463134399e-05,
+      "loss": 1.262,
+      "step": 280
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 4.780094502636552e-05,
+      "loss": 1.255,
+      "step": 290
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 4.764913428413572e-05,
+      "loss": 1.2652,
+      "step": 300
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 4.7492514637381727e-05,
+      "loss": 1.2668,
+      "step": 310
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 4.733111933986583e-05,
+      "loss": 1.2621,
+      "step": 320
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 4.716498265932501e-05,
+      "loss": 1.257,
+      "step": 330
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 4.699413987019512e-05,
+      "loss": 1.2789,
+      "step": 340
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 4.681862724612141e-05,
+      "loss": 1.2634,
+      "step": 350
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 4.663848205225674e-05,
+      "loss": 1.2594,
+      "step": 360
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 4.645374253734949e-05,
+      "loss": 1.26,
+      "step": 370
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 4.626444792562244e-05,
+      "loss": 1.2514,
+      "step": 380
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 4.607063840844463e-05,
+      "loss": 1.2506,
+      "step": 390
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 4.587235513579791e-05,
+      "loss": 1.2648,
+      "step": 400
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 4.5669640207539786e-05,
+      "loss": 1.2511,
+      "step": 410
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 4.546253666446484e-05,
+      "loss": 1.2594,
+      "step": 420
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 4.525108847916614e-05,
+      "loss": 1.2608,
+      "step": 430
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 4.503534054669892e-05,
+      "loss": 1.2597,
+      "step": 440
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 4.481533867504841e-05,
+      "loss": 1.2609,
+      "step": 450
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 4.4591129575403765e-05,
+      "loss": 1.2505,
+      "step": 460
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 4.43627608522403e-05,
+      "loss": 1.2481,
+      "step": 470
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 4.4130280993211974e-05,
+      "loss": 1.2612,
+      "step": 480
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 4.389373935885646e-05,
+      "loss": 1.2504,
+      "step": 490
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 4.365318617211479e-05,
+      "loss": 1.2518,
+      "step": 500
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 4.340867250766794e-05,
+      "loss": 1.2458,
+      "step": 510
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 4.316025028109258e-05,
+      "loss": 1.2345,
+      "step": 520
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 4.2907972237838225e-05,
+      "loss": 1.2521,
+      "step": 530
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 4.2651891942028274e-05,
+      "loss": 1.2528,
+      "step": 540
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 4.239206376508717e-05,
+      "loss": 1.2462,
+      "step": 550
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 4.212854287419611e-05,
+      "loss": 1.2401,
+      "step": 560
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 4.1861385220579934e-05,
+      "loss": 1.2496,
+      "step": 570
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 4.1590647527627404e-05,
+      "loss": 1.2522,
+      "step": 580
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 4.131638727884762e-05,
+      "loss": 1.2377,
+      "step": 590
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 4.103866270566498e-05,
+      "loss": 1.2467,
+      "step": 600
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 4.075753277505544e-05,
+      "loss": 1.2421,
+      "step": 610
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 4.0473057177026484e-05,
+      "loss": 1.2455,
+      "step": 620
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 4.018529631194369e-05,
+      "loss": 1.2294,
+      "step": 630
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 3.989431127770635e-05,
+      "loss": 1.2509,
+      "step": 640
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 3.960016385677513e-05,
+      "loss": 1.2354,
+      "step": 650
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 3.9302916503054246e-05,
+      "loss": 1.2318,
+      "step": 660
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 3.9002632328631164e-05,
+      "loss": 1.2376,
+      "step": 670
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 3.8699375090376534e-05,
+      "loss": 1.2412,
+      "step": 680
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 3.8393209176407223e-05,
+      "loss": 1.2479,
+      "step": 690
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 3.8084199592415305e-05,
+      "loss": 1.2431,
+      "step": 700
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 3.777241194786591e-05,
+      "loss": 1.245,
+      "step": 710
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 3.745791244206697e-05,
+      "loss": 1.2393,
+      "step": 720
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 3.714076785011359e-05,
+      "loss": 1.2473,
+      "step": 730
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 3.682104550871031e-05,
+      "loss": 1.2552,
+      "step": 740
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 3.649881330187401e-05,
+      "loss": 1.2356,
+      "step": 750
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 3.617413964652067e-05,
+      "loss": 1.2442,
+      "step": 760
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 3.5847093477938956e-05,
+      "loss": 1.2605,
+      "step": 770
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 3.551774423515378e-05,
+      "loss": 1.2364,
+      "step": 780
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 3.518616184618288e-05,
+      "loss": 1.2482,
+      "step": 790
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 3.4852416713189526e-05,
+      "loss": 1.2327,
+      "step": 800
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 3.4516579697534705e-05,
+      "loss": 1.2548,
+      "step": 810
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 3.417872210473162e-05,
+      "loss": 1.2469,
+      "step": 820
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 3.3838915669306034e-05,
+      "loss": 1.2424,
+      "step": 830
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 3.349723253956542e-05,
+      "loss": 1.2438,
+      "step": 840
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 3.315374526228036e-05,
+      "loss": 1.243,
+      "step": 850
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 3.2808526767281225e-05,
+      "loss": 1.2416,
+      "step": 860
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 3.246165035197364e-05,
+      "loss": 1.2399,
+      "step": 870
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 3.211318966577581e-05,
+      "loss": 1.239,
+      "step": 880
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 3.176321869448116e-05,
+      "loss": 1.2498,
+      "step": 890
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 3.1411811744549536e-05,
+      "loss": 1.2438,
+      "step": 900
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 3.105904342733032e-05,
+      "loss": 1.233,
+      "step": 910
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 3.070498864322081e-05,
+      "loss": 1.2315,
+      "step": 920
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 3.034972256576328e-05,
+      "loss": 1.232,
+      "step": 930
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 2.999332062568395e-05,
+      "loss": 1.2467,
+      "step": 940
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 2.9635858494877384e-05,
+      "loss": 1.2483,
+      "step": 950
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 2.9277412070339782e-05,
+      "loss": 1.2377,
+      "step": 960
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 2.891805745805429e-05,
+      "loss": 1.2347,
+      "step": 970
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 2.8557870956832132e-05,
+      "loss": 1.2428,
+      "step": 980
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 2.8196929042112652e-05,
+      "loss": 1.244,
+      "step": 990
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 2.783530834972594e-05,
+      "loss": 1.2317,
+      "step": 1000
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 2.7473085659621377e-05,
+      "loss": 1.2421,
+      "step": 1010
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 2.711033787956555e-05,
+      "loss": 1.2446,
+      "step": 1020
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 2.6747142028813105e-05,
+      "loss": 1.2471,
+      "step": 1030
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 2.638357522175383e-05,
+      "loss": 1.2403,
+      "step": 1040
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 2.6019714651539646e-05,
+      "loss": 1.2413,
+      "step": 1050
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 2.565563757369475e-05,
+      "loss": 1.2341,
+      "step": 1060
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 2.529142128971268e-05,
+      "loss": 1.236,
+      "step": 1070
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.492714313064342e-05,
+      "loss": 1.2423,
+      "step": 1080
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.45628804406744e-05,
+      "loss": 1.2253,
+      "step": 1090
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.419871056070862e-05,
+      "loss": 1.2377,
+      "step": 1100
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 2.3834710811943514e-05,
+      "loss": 1.2395,
+      "step": 1110
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 2.3470958479453938e-05,
+      "loss": 1.2238,
+      "step": 1120
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 2.3107530795782877e-05,
+      "loss": 1.2045,
+      "step": 1130
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 2.2744504924543313e-05,
+      "loss": 1.2302,
+      "step": 1140
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 2.23819579440347e-05,
+      "loss": 1.233,
+      "step": 1150
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 2.2019966830877545e-05,
+      "loss": 1.2341,
+      "step": 1160
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 2.1658608443669635e-05,
+      "loss": 1.2294,
+      "step": 1170
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 2.1297959506667224e-05,
+      "loss": 1.2308,
+      "step": 1180
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 2.0938096593494855e-05,
+      "loss": 1.2427,
+      "step": 1190
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 2.057909611088709e-05,
+      "loss": 1.2404,
+      "step": 1200
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 2.02210342824657e-05,
+      "loss": 1.2245,
+      "step": 1210
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.9863987132555706e-05,
+      "loss": 1.2288,
+      "step": 1220
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.9508030470043806e-05,
+      "loss": 1.2401,
+      "step": 1230
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 1.915323987228247e-05,
+      "loss": 1.2275,
+      "step": 1240
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.8799690669043212e-05,
+      "loss": 1.2251,
+      "step": 1250
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 1.8447457926522454e-05,
+      "loss": 1.238,
+      "step": 1260
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.8096616431403325e-05,
+      "loss": 1.2291,
+      "step": 1270
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.7747240674976857e-05,
+      "loss": 1.2162,
+      "step": 1280
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.7399404837325796e-05,
+      "loss": 1.2315,
+      "step": 1290
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.7053182771574633e-05,
+      "loss": 1.2383,
+      "step": 1300
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.6708647988208887e-05,
+      "loss": 1.2254,
+      "step": 1310
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.6365873639467315e-05,
+      "loss": 1.2388,
+      "step": 1320
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.602493250381003e-05,
+      "loss": 1.2213,
+      "step": 1330
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.5685896970466123e-05,
+      "loss": 1.2366,
+      "step": 1340
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 1.534883902406375e-05,
+      "loss": 1.2265,
+      "step": 1350
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.5013830229346326e-05,
+      "loss": 1.2094,
+      "step": 1360
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.4680941715977722e-05,
+      "loss": 1.2314,
+      "step": 1370
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 1.4350244163439892e-05,
+      "loss": 1.2254,
+      "step": 1380
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.4021807786026108e-05,
+      "loss": 1.2287,
+      "step": 1390
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.3695702317932862e-05,
+      "loss": 1.2235,
+      "step": 1400
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.337199699845387e-05,
+      "loss": 1.2209,
+      "step": 1410
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.3050760557279023e-05,
+      "loss": 1.2296,
+      "step": 1420
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 1.2732061199901562e-05,
+      "loss": 1.2363,
+      "step": 1430
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.2415966593136547e-05,
+      "loss": 1.2116,
+      "step": 1440
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.2102543850753808e-05,
+      "loss": 1.2299,
+      "step": 1450
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.1791859519228138e-05,
+      "loss": 1.2204,
+      "step": 1460
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 1.148397956361007e-05,
+      "loss": 1.2271,
+      "step": 1470
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 1.1178969353520018e-05,
+      "loss": 1.2327,
+      "step": 1480
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.087689364926897e-05,
+      "loss": 1.2252,
+      "step": 1490
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.0577816588108378e-05,
+      "loss": 1.2305,
+      "step": 1500
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 1.0281801670612449e-05,
+      "loss": 1.2179,
+      "step": 1510
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 9.988911747195603e-06,
+      "loss": 1.229,
+      "step": 1520
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 9.699209004767953e-06,
+      "loss": 1.2192,
+      "step": 1530
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 9.412754953531663e-06,
+      "loss": 1.2322,
+      "step": 1540
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 9.129610413921005e-06,
+      "loss": 1.2193,
+      "step": 1550
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 8.849835503688847e-06,
+      "loss": 1.2149,
+      "step": 1560
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 8.57348962514234e-06,
+      "loss": 1.217,
+      "step": 1570
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 8.30063145253053e-06,
+      "loss": 1.2368,
+      "step": 1580
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 8.031318919586523e-06,
+      "loss": 1.2264,
+      "step": 1590
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 7.7656092072269e-06,
+      "loss": 1.2243,
+      "step": 1600
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 7.503558731410959e-06,
+      "loss": 1.2427,
+      "step": 1610
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 7.245223131162376e-06,
+      "loss": 1.2207,
+      "step": 1620
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 6.9906572567558285e-06,
+      "loss": 1.2211,
+      "step": 1630
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 6.739915158071106e-06,
+      "loss": 1.2343,
+      "step": 1640
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 6.493050073117116e-06,
+      "loss": 1.2348,
+      "step": 1650
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 6.250114416728298e-06,
+      "loss": 1.2189,
+      "step": 1660
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 6.011159769435823e-06,
+      "loss": 1.2207,
+      "step": 1670
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 5.776236866515947e-06,
+      "loss": 1.2315,
+      "step": 1680
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 5.545395587217763e-06,
+      "loss": 1.2334,
+      "step": 1690
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 5.318684944172752e-06,
+      "loss": 1.232,
+      "step": 1700
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 5.096153072988366e-06,
+      "loss": 1.2235,
+      "step": 1710
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 4.8778472220277874e-06,
+      "loss": 1.2244,
+      "step": 1720
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 4.6638137423780395e-06,
+      "loss": 1.2197,
+      "step": 1730
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 4.454098078008667e-06,
+      "loss": 1.2242,
+      "step": 1740
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 4.248744756122986e-06,
+      "loss": 1.2382,
+      "step": 1750
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 4.047797377703985e-06,
+      "loss": 1.2265,
+      "step": 1760
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 3.851298608256892e-06,
+      "loss": 1.2153,
+      "step": 1770
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 3.6592901687503566e-06,
+      "loss": 1.2198,
+      "step": 1780
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 3.471812826758178e-06,
+      "loss": 1.23,
+      "step": 1790
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 3.288906387803464e-06,
+      "loss": 1.2268,
+      "step": 1800
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 3.1106096869070483e-06,
+      "loss": 1.2228,
+      "step": 1810
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 2.9369605803419715e-06,
+      "loss": 1.2288,
+      "step": 1820
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 2.767995937595766e-06,
+      "loss": 1.228,
+      "step": 1830
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 2.6037516335422728e-06,
+      "loss": 1.2178,
+      "step": 1840
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 2.4442625408246074e-06,
+      "loss": 1.2386,
+      "step": 1850
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 2.289562522450947e-06,
+      "loss": 1.2274,
+      "step": 1860
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 2.1396844246046903e-06,
+      "loss": 1.2341,
+      "step": 1870
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 1.9946600696704592e-06,
+      "loss": 1.2118,
+      "step": 1880
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 1.8545202494775509e-06,
+      "loss": 1.2219,
+      "step": 1890
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 1.7192947187621434e-06,
+      "loss": 1.2268,
+      "step": 1900
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 1.5890121888497366e-06,
+      "loss": 1.2183,
+      "step": 1910
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 1.463700321559075e-06,
+      "loss": 1.2253,
+      "step": 1920
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 1.3433857233289714e-06,
+      "loss": 1.2262,
+      "step": 1930
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 1.2280939395691859e-06,
+      "loss": 1.2232,
+      "step": 1940
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 1.1178494492365465e-06,
+      "loss": 1.2376,
+      "step": 1950
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 1.0126756596375686e-06,
+      "loss": 1.2287,
+      "step": 1960
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 9.125949014585383e-07,
+      "loss": 1.2354,
+      "step": 1970
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 8.176284240242638e-07,
+      "loss": 1.224,
+      "step": 1980
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 7.277963907863478e-07,
+      "loss": 1.2214,
+      "step": 1990
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 6.431178750420513e-07,
+      "loss": 1.2214,
+      "step": 2000
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 2156,
+  "num_train_epochs": 2,
+  "save_steps": 1000,
+  "total_flos": 3.1404803520494305e+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:e1ca0e4372bd1720a48da585ce6946399a06d0c984dcfcf5a2b7104827c4603f
+size 4219

special_tokens_map.json

@@ -0,0 +1,7 @@
+{
+  "additional_special_tokens": [
+    "<|im_end|>"
+  ],
+  "eos_token": "<|endoftext|>",
+  "pad_token": "<|endoftext|>"
+}

tokenization_qwen.py

@@ -0,0 +1,246 @@
+# Copyright (c) Alibaba Cloud.
+#
+# This source code is licensed under the license found in the
+# LICENSE file in the root directory of this source tree.
+
+"""Tokenization classes for QWen."""
+
+import base64
+import logging
+import os
+import unicodedata
+from typing import Collection, Dict, List, Set, Tuple, Union
+
+import tiktoken
+from transformers import PreTrainedTokenizer, AddedToken
+
+logger = logging.getLogger(__name__)
+
+
+VOCAB_FILES_NAMES = {"vocab_file": "qwen.tiktoken"}
+
+PAT_STR = r"""(?i:'s|'t|'re|'ve|'m|'ll|'d)|[^\r\n\p{L}\p{N}]?\p{L}+|\p{N}| ?[^\s\p{L}\p{N}]+[\r\n]*|\s*[\r\n]+|\s+(?!\S)|\s+"""
+ENDOFTEXT = "<|endoftext|>"
+IMSTART = "<|im_start|>"
+IMEND = "<|im_end|>"
+# as the default behavior is changed to allow special tokens in
+# regular texts, the surface forms of special tokens need to be
+# as different as possible to minimize the impact
+EXTRAS = tuple((f"<|extra_{i}|>" for i in range(205)))
+SPECIAL_TOKENS = (
+    ENDOFTEXT,
+    IMSTART,
+    IMEND,
+) + EXTRAS
+
+
+def _load_tiktoken_bpe(tiktoken_bpe_file: str) -> Dict[bytes, int]:
+    with open(tiktoken_bpe_file, "rb") as f:
+        contents = f.read()
+    return {
+        base64.b64decode(token): int(rank)
+        for token, rank in (line.split() for line in contents.splitlines() if line)
+    }
+
+class QWenTokenizer(PreTrainedTokenizer):
+    """QWen tokenizer."""
+
+    vocab_files_names = VOCAB_FILES_NAMES
+
+    def __init__(
+        self,
+        vocab_file,
+        errors="replace",
+        **kwargs,
+    ):
+        super().__init__(**kwargs)
+
+        self.errors = errors  # how to handle errors in decoding
+
+        self.mergeable_ranks = _load_tiktoken_bpe(vocab_file)  # type: dict[bytes, int]
+        self.special_tokens = {
+            token: index
+            for index, token in enumerate(
+                SPECIAL_TOKENS, start=len(self.mergeable_ranks)
+            )
+        }
+
+        enc = tiktoken.Encoding(
+            "Qwen",
+            pat_str=PAT_STR,
+            mergeable_ranks=self.mergeable_ranks,
+            special_tokens=self.special_tokens,
+        )
+        assert (
+            len(self.mergeable_ranks) + len(self.special_tokens) == enc.n_vocab
+        ), f"{len(self.mergeable_ranks) + len(self.special_tokens)} != {enc.n_vocab} in encoding"
+
+        self.decoder = {
+            v: k for k, v in self.mergeable_ranks.items()
+        }  # type: dict[int, bytes|str]
+        self.decoder.update({v: k for k, v in self.special_tokens.items()})
+
+        self.tokenizer = enc  # type: tiktoken.Encoding
+
+        self.eod_id = self.tokenizer.eot_token
+        self.im_start_id = self.special_tokens[IMSTART]
+        self.im_end_id = self.special_tokens[IMEND]
+
+    def __getstate__(self):
+        # for pickle lovers
+        state = self.__dict__.copy()
+        del state['tokenizer']
+        return state
+
+    def __setstate__(self, state):
+        # tokenizer is not python native; don't pass it; rebuild it
+        self.__dict__.update(state)
+        enc = tiktoken.Encoding(
+            "Qwen",
+            pat_str=PAT_STR,
+            mergeable_ranks=self.mergeable_ranks,
+            special_tokens=self.special_tokens,
+        )
+        self.tokenizer = enc
+
+
+    def __len__(self) -> int:
+        return self.tokenizer.n_vocab
+
+    def get_vocab(self) -> Dict[bytes, int]:
+        return self.mergeable_ranks
+
+    def convert_tokens_to_ids(
+        self, tokens: Union[bytes, str, List[Union[bytes, str]]]
+    ) -> List[int]:
+        ids = []
+        if isinstance(tokens, (str, bytes)):
+            if tokens in self.special_tokens:
+                return self.special_tokens[tokens]
+            else:
+                return self.mergeable_ranks.get(tokens)
+        for token in tokens:
+            if token in self.special_tokens:
+                ids.append(self.special_tokens[token])
+            else:
+                ids.append(self.mergeable_ranks.get(token))
+        return ids
+
+    def _add_tokens(self, new_tokens: Union[List[str], List[AddedToken]], special_tokens: bool = False) -> int:
+        if not special_tokens and new_tokens:
+            raise ValueError('Adding regular tokens is not supported')
+        for token in new_tokens:
+            surface_form = token.content if isinstance(token, AddedToken) else token
+            if surface_form not in SPECIAL_TOKENS:
+                raise ValueError('Adding unknown special tokens is not supported')
+        return 0
+
+    def save_vocabulary(self, save_directory: str, **kwargs) -> Tuple[str]:
+        """
+        Save only the vocabulary of the tokenizer (vocabulary).
+
+        Returns:
+            `Tuple(str)`: Paths to the files saved.
+        """
+        file_path = os.path.join(save_directory, "qwen.tiktoken")
+        with open(file_path, "w", encoding="utf8") as w:
+            for k, v in self.mergeable_ranks.items():
+                line = base64.b64encode(k).decode("utf8") + " " + str(v) + "\n"
+                w.write(line)
+        return (file_path,)
+
+    def tokenize(
+        self,
+        text: str,
+        allowed_special: Union[Set, str] = "all",
+        disallowed_special: Union[Collection, str] = (),
+        **kwargs,
+    ) -> List[Union[bytes, str]]:
+        """
+        Converts a string in a sequence of tokens.
+
+        Args:
+            text (`str`):
+                The sequence to be encoded.
+            allowed_special (`Literal["all"]` or `set`):
+                The surface forms of the tokens to be encoded as special tokens in regular texts.
+                Default to "all".
+            disallowed_special (`Literal["all"]` or `Collection`):
+                The surface forms of the tokens that should not be in regular texts and trigger errors.
+                Default to an empty tuple.
+
+            kwargs (additional keyword arguments, *optional*):
+                Will be passed to the underlying model specific encode method.
+
+        Returns:
+            `List[bytes|str]`: The list of tokens.
+        """
+        tokens = []
+        text = unicodedata.normalize("NFC", text)
+
+        # this implementation takes a detour: text -> token id -> token surface forms
+        for t in self.tokenizer.encode(
+            text, allowed_special=allowed_special, disallowed_special=disallowed_special
+        ):
+            tokens.append(self.decoder[t])
+        return tokens
+
+    def convert_tokens_to_string(self, tokens: List[Union[bytes, str]]) -> str:
+        """
+        Converts a sequence of tokens in a single string.
+        """
+        text = ""
+        temp = b""
+        for t in tokens:
+            if isinstance(t, str):
+                if temp:
+                    text += temp.decode("utf-8", errors=self.errors)
+                    temp = b""
+                text += t
+            elif isinstance(t, bytes):
+                temp += t
+            else:
+                raise TypeError("token should only be of type types or str")
+        if temp:
+            text += temp.decode("utf-8", errors=self.errors)
+        return text
+
+    @property
+    def vocab_size(self):
+        return self.tokenizer.n_vocab
+
+    def _convert_id_to_token(self, index: int) -> Union[bytes, str]:
+        """Converts an id to a token, special tokens included"""
+        if index in self.decoder:
+            return self.decoder[index]
+        raise ValueError("unknown ids")
+
+    def _convert_token_to_id(self, token: Union[bytes, str]) -> int:
+        """Converts a token to an id using the vocab, special tokens included"""
+        if token in self.special_tokens:
+            return self.special_tokens[token]
+        if token in self.mergeable_ranks:
+            return self.mergeable_ranks[token]
+        raise ValueError("unknown token")
+
+    def _tokenize(self, text: str, **kwargs):
+        """
+        Converts a string in a sequence of tokens (string), using the tokenizer. Split in words for word-based
+        vocabulary or sub-words for sub-word-based vocabularies (BPE/SentencePieces/WordPieces).
+
+        Do NOT take care of added tokens.
+        """
+        raise NotImplementedError
+
+    def _decode(
+        self,
+        token_ids: Union[int, List[int]],
+        skip_special_tokens: bool = False,
+        errors: str = None,
+        **kwargs,
+    ) -> str:
+        if isinstance(token_ids, int):
+            token_ids = [token_ids]
+        if skip_special_tokens:
+            token_ids = [i for i in token_ids if i < self.eod_id]
+        return self.tokenizer.decode(token_ids, errors=errors or self.errors)

tokenizer_config.json

@@ -0,0 +1,12 @@
+{
+  "auto_map": {
+    "AutoTokenizer": [
+      "tokenization_qwen.QWenTokenizer",
+      null
+    ]
+  },
+  "clean_up_tokenization_spaces": true,
+  "model_max_length": 8192,
+  "padding_side": "right",
+  "tokenizer_class": "QWenTokenizer"
+}

train_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 2.0,
+    "train_loss": 1.2461530792912217,
+    "train_runtime": 30560.3748,
+    "train_samples_per_second": 4.517,
+    "train_steps_per_second": 0.071
+}

trainer_log.jsonl

@@ -0,0 +1,216 @@
+{"current_steps": 10, "total_steps": 2156, "loss": 1.6199, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.999734597774032e-05, "epoch": 0.01, "percentage": 0.46, "elapsed_time": "0:02:25", "remaining_time": "8:39:45"}
+{"current_steps": 20, "total_steps": 2156, "loss": 1.5062, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.998938447446803e-05, "epoch": 0.02, "percentage": 0.93, "elapsed_time": "0:04:50", "remaining_time": "8:37:49"}
+{"current_steps": 30, "total_steps": 2156, "loss": 1.4138, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.997611718058365e-05, "epoch": 0.03, "percentage": 1.39, "elapsed_time": "0:07:09", "remaining_time": "8:26:44"}
+{"current_steps": 40, "total_steps": 2156, "loss": 1.369, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.9957546913022665e-05, "epoch": 0.04, "percentage": 1.86, "elapsed_time": "0:09:34", "remaining_time": "8:26:05"}
+{"current_steps": 50, "total_steps": 2156, "loss": 1.3408, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.993367761465736e-05, "epoch": 0.05, "percentage": 2.32, "elapsed_time": "0:11:55", "remaining_time": "8:21:57"}
+{"current_steps": 60, "total_steps": 2156, "loss": 1.3321, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.9904514353459654e-05, "epoch": 0.06, "percentage": 2.78, "elapsed_time": "0:14:18", "remaining_time": "8:20:02"}
+{"current_steps": 70, "total_steps": 2156, "loss": 1.3251, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.9870063321425105e-05, "epoch": 0.06, "percentage": 3.25, "elapsed_time": "0:16:39", "remaining_time": "8:16:26"}
+{"current_steps": 80, "total_steps": 2156, "loss": 1.3228, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.983033183325818e-05, "epoch": 0.07, "percentage": 3.71, "elapsed_time": "0:18:55", "remaining_time": "8:11:08"}
+{"current_steps": 90, "total_steps": 2156, "loss": 1.3111, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.97853283248192e-05, "epoch": 0.08, "percentage": 4.17, "elapsed_time": "0:21:14", "remaining_time": "8:07:25"}
+{"current_steps": 100, "total_steps": 2156, "loss": 1.3013, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.973506235133323e-05, "epoch": 0.09, "percentage": 4.64, "elapsed_time": "0:23:37", "remaining_time": "8:05:34"}
+{"current_steps": 110, "total_steps": 2156, "loss": 1.3004, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.967954458536126e-05, "epoch": 0.1, "percentage": 5.1, "elapsed_time": "0:26:01", "remaining_time": "8:03:58"}
+{"current_steps": 120, "total_steps": 2156, "loss": 1.2959, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.9618786814534226e-05, "epoch": 0.11, "percentage": 5.57, "elapsed_time": "0:28:23", "remaining_time": "8:01:40"}
+{"current_steps": 130, "total_steps": 2156, "loss": 1.2969, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.955280193905022e-05, "epoch": 0.12, "percentage": 6.03, "elapsed_time": "0:30:48", "remaining_time": "8:00:05"}
+{"current_steps": 140, "total_steps": 2156, "loss": 1.2879, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.948160396893553e-05, "epoch": 0.13, "percentage": 6.49, "elapsed_time": "0:33:11", "remaining_time": "7:57:58"}
+{"current_steps": 150, "total_steps": 2156, "loss": 1.277, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.9405208021069946e-05, "epoch": 0.14, "percentage": 6.96, "elapsed_time": "0:35:30", "remaining_time": "7:54:48"}
+{"current_steps": 160, "total_steps": 2156, "loss": 1.283, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.9323630315977156e-05, "epoch": 0.15, "percentage": 7.42, "elapsed_time": "0:37:55", "remaining_time": "7:53:08"}
+{"current_steps": 170, "total_steps": 2156, "loss": 1.288, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.9236888174380784e-05, "epoch": 0.16, "percentage": 7.88, "elapsed_time": "0:40:13", "remaining_time": "7:49:57"}
+{"current_steps": 180, "total_steps": 2156, "loss": 1.287, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.91450000135268e-05, "epoch": 0.17, "percentage": 8.35, "elapsed_time": "0:42:34", "remaining_time": "7:47:18"}
+{"current_steps": 190, "total_steps": 2156, "loss": 1.2726, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.9047985343273154e-05, "epoch": 0.18, "percentage": 8.81, "elapsed_time": "0:44:50", "remaining_time": "7:43:57"}
+{"current_steps": 200, "total_steps": 2156, "loss": 1.2808, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.894586476194739e-05, "epoch": 0.19, "percentage": 9.28, "elapsed_time": "0:47:07", "remaining_time": "7:40:53"}
+{"current_steps": 210, "total_steps": 2156, "loss": 1.2657, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.883865995197319e-05, "epoch": 0.19, "percentage": 9.74, "elapsed_time": "0:49:29", "remaining_time": "7:38:40"}
+{"current_steps": 220, "total_steps": 2156, "loss": 1.275, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.8726393675266716e-05, "epoch": 0.2, "percentage": 10.2, "elapsed_time": "0:51:55", "remaining_time": "7:36:59"}
+{"current_steps": 230, "total_steps": 2156, "loss": 1.2667, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.860908976840376e-05, "epoch": 0.21, "percentage": 10.67, "elapsed_time": "0:54:19", "remaining_time": "7:34:51"}
+{"current_steps": 240, "total_steps": 2156, "loss": 1.2715, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.848677313755872e-05, "epoch": 0.22, "percentage": 11.13, "elapsed_time": "0:56:35", "remaining_time": "7:31:45"}
+{"current_steps": 250, "total_steps": 2156, "loss": 1.273, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.835946975321647e-05, "epoch": 0.23, "percentage": 11.6, "elapsed_time": "0:58:54", "remaining_time": "7:29:08"}
+{"current_steps": 260, "total_steps": 2156, "loss": 1.2605, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.822720664465827e-05, "epoch": 0.24, "percentage": 12.06, "elapsed_time": "1:01:17", "remaining_time": "7:26:57"}
+{"current_steps": 270, "total_steps": 2156, "loss": 1.2766, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.809001189422287e-05, "epoch": 0.25, "percentage": 12.52, "elapsed_time": "1:03:39", "remaining_time": "7:24:39"}
+{"current_steps": 280, "total_steps": 2156, "loss": 1.262, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.794791463134399e-05, "epoch": 0.26, "percentage": 12.99, "elapsed_time": "1:05:57", "remaining_time": "7:21:56"}
+{"current_steps": 290, "total_steps": 2156, "loss": 1.255, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.780094502636552e-05, "epoch": 0.27, "percentage": 13.45, "elapsed_time": "1:08:16", "remaining_time": "7:19:18"}
+{"current_steps": 300, "total_steps": 2156, "loss": 1.2652, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.764913428413572e-05, "epoch": 0.28, "percentage": 13.91, "elapsed_time": "1:10:42", "remaining_time": "7:17:25"}
+{"current_steps": 310, "total_steps": 2156, "loss": 1.2668, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.7492514637381727e-05, "epoch": 0.29, "percentage": 14.38, "elapsed_time": "1:13:01", "remaining_time": "7:14:52"}
+{"current_steps": 320, "total_steps": 2156, "loss": 1.2621, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.733111933986583e-05, "epoch": 0.3, "percentage": 14.84, "elapsed_time": "1:15:23", "remaining_time": "7:12:34"}
+{"current_steps": 330, "total_steps": 2156, "loss": 1.257, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.716498265932501e-05, "epoch": 0.31, "percentage": 15.31, "elapsed_time": "1:17:43", "remaining_time": "7:10:04"}
+{"current_steps": 340, "total_steps": 2156, "loss": 1.2789, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.699413987019512e-05, "epoch": 0.32, "percentage": 15.77, "elapsed_time": "1:20:02", "remaining_time": "7:07:30"}
+{"current_steps": 350, "total_steps": 2156, "loss": 1.2634, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.681862724612141e-05, "epoch": 0.32, "percentage": 16.23, "elapsed_time": "1:22:26", "remaining_time": "7:05:24"}
+{"current_steps": 360, "total_steps": 2156, "loss": 1.2594, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.663848205225674e-05, "epoch": 0.33, "percentage": 16.7, "elapsed_time": "1:24:46", "remaining_time": "7:02:55"}
+{"current_steps": 370, "total_steps": 2156, "loss": 1.26, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.645374253734949e-05, "epoch": 0.34, "percentage": 17.16, "elapsed_time": "1:27:06", "remaining_time": "7:00:28"}
+{"current_steps": 380, "total_steps": 2156, "loss": 1.2514, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.626444792562244e-05, "epoch": 0.35, "percentage": 17.63, "elapsed_time": "1:29:26", "remaining_time": "6:58:02"}
+{"current_steps": 390, "total_steps": 2156, "loss": 1.2506, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.607063840844463e-05, "epoch": 0.36, "percentage": 18.09, "elapsed_time": "1:31:45", "remaining_time": "6:55:30"}
+{"current_steps": 400, "total_steps": 2156, "loss": 1.2648, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.587235513579791e-05, "epoch": 0.37, "percentage": 18.55, "elapsed_time": "1:34:04", "remaining_time": "6:53:00"}
+{"current_steps": 410, "total_steps": 2156, "loss": 1.2511, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.5669640207539786e-05, "epoch": 0.38, "percentage": 19.02, "elapsed_time": "1:36:25", "remaining_time": "6:50:36"}
+{"current_steps": 420, "total_steps": 2156, "loss": 1.2594, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.546253666446484e-05, "epoch": 0.39, "percentage": 19.48, "elapsed_time": "1:38:47", "remaining_time": "6:48:20"}
+{"current_steps": 430, "total_steps": 2156, "loss": 1.2608, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.525108847916614e-05, "epoch": 0.4, "percentage": 19.94, "elapsed_time": "1:41:06", "remaining_time": "6:45:52"}
+{"current_steps": 440, "total_steps": 2156, "loss": 1.2597, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.503534054669892e-05, "epoch": 0.41, "percentage": 20.41, "elapsed_time": "1:43:28", "remaining_time": "6:43:33"}
+{"current_steps": 450, "total_steps": 2156, "loss": 1.2609, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.481533867504841e-05, "epoch": 0.42, "percentage": 20.87, "elapsed_time": "1:45:51", "remaining_time": "6:41:19"}
+{"current_steps": 460, "total_steps": 2156, "loss": 1.2505, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.4591129575403765e-05, "epoch": 0.43, "percentage": 21.34, "elapsed_time": "1:48:11", "remaining_time": "6:38:53"}
+{"current_steps": 470, "total_steps": 2156, "loss": 1.2481, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.43627608522403e-05, "epoch": 0.44, "percentage": 21.8, "elapsed_time": "1:50:39", "remaining_time": "6:36:56"}
+{"current_steps": 480, "total_steps": 2156, "loss": 1.2612, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.4130280993211974e-05, "epoch": 0.45, "percentage": 22.26, "elapsed_time": "1:53:04", "remaining_time": "6:34:50"}
+{"current_steps": 490, "total_steps": 2156, "loss": 1.2504, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.389373935885646e-05, "epoch": 0.45, "percentage": 22.73, "elapsed_time": "1:55:26", "remaining_time": "6:32:29"}
+{"current_steps": 500, "total_steps": 2156, "loss": 1.2518, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.365318617211479e-05, "epoch": 0.46, "percentage": 23.19, "elapsed_time": "1:57:51", "remaining_time": "6:30:21"}
+{"current_steps": 510, "total_steps": 2156, "loss": 1.2458, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.340867250766794e-05, "epoch": 0.47, "percentage": 23.65, "elapsed_time": "2:00:13", "remaining_time": "6:28:02"}
+{"current_steps": 520, "total_steps": 2156, "loss": 1.2345, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.316025028109258e-05, "epoch": 0.48, "percentage": 24.12, "elapsed_time": "2:02:34", "remaining_time": "6:25:37"}
+{"current_steps": 530, "total_steps": 2156, "loss": 1.2521, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.2907972237838225e-05, "epoch": 0.49, "percentage": 24.58, "elapsed_time": "2:04:58", "remaining_time": "6:23:23"}
+{"current_steps": 540, "total_steps": 2156, "loss": 1.2528, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.2651891942028274e-05, "epoch": 0.5, "percentage": 25.05, "elapsed_time": "2:07:19", "remaining_time": "6:21:02"}
+{"current_steps": 550, "total_steps": 2156, "loss": 1.2462, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.239206376508717e-05, "epoch": 0.51, "percentage": 25.51, "elapsed_time": "2:09:41", "remaining_time": "6:18:42"}
+{"current_steps": 560, "total_steps": 2156, "loss": 1.2401, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.212854287419611e-05, "epoch": 0.52, "percentage": 25.97, "elapsed_time": "2:12:06", "remaining_time": "6:16:31"}
+{"current_steps": 570, "total_steps": 2156, "loss": 1.2496, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.1861385220579934e-05, "epoch": 0.53, "percentage": 26.44, "elapsed_time": "2:14:29", "remaining_time": "6:14:13"}
+{"current_steps": 580, "total_steps": 2156, "loss": 1.2522, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.1590647527627404e-05, "epoch": 0.54, "percentage": 26.9, "elapsed_time": "2:16:49", "remaining_time": "6:11:48"}
+{"current_steps": 590, "total_steps": 2156, "loss": 1.2377, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.131638727884762e-05, "epoch": 0.55, "percentage": 27.37, "elapsed_time": "2:19:12", "remaining_time": "6:09:29"}
+{"current_steps": 600, "total_steps": 2156, "loss": 1.2467, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.103866270566498e-05, "epoch": 0.56, "percentage": 27.83, "elapsed_time": "2:21:36", "remaining_time": "6:07:13"}
+{"current_steps": 610, "total_steps": 2156, "loss": 1.2421, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.075753277505544e-05, "epoch": 0.57, "percentage": 28.29, "elapsed_time": "2:23:53", "remaining_time": "6:04:41"}
+{"current_steps": 620, "total_steps": 2156, "loss": 1.2455, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.0473057177026484e-05, "epoch": 0.57, "percentage": 28.76, "elapsed_time": "2:26:14", "remaining_time": "6:02:18"}
+{"current_steps": 630, "total_steps": 2156, "loss": 1.2294, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.018529631194369e-05, "epoch": 0.58, "percentage": 29.22, "elapsed_time": "2:28:37", "remaining_time": "5:59:59"}
+{"current_steps": 640, "total_steps": 2156, "loss": 1.2509, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.989431127770635e-05, "epoch": 0.59, "percentage": 29.68, "elapsed_time": "2:31:04", "remaining_time": "5:57:50"}
+{"current_steps": 650, "total_steps": 2156, "loss": 1.2354, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.960016385677513e-05, "epoch": 0.6, "percentage": 30.15, "elapsed_time": "2:33:22", "remaining_time": "5:55:20"}
+{"current_steps": 660, "total_steps": 2156, "loss": 1.2318, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.9302916503054246e-05, "epoch": 0.61, "percentage": 30.61, "elapsed_time": "2:35:43", "remaining_time": "5:52:59"}
+{"current_steps": 670, "total_steps": 2156, "loss": 1.2376, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.9002632328631164e-05, "epoch": 0.62, "percentage": 31.08, "elapsed_time": "2:38:03", "remaining_time": "5:50:33"}
+{"current_steps": 680, "total_steps": 2156, "loss": 1.2412, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.8699375090376534e-05, "epoch": 0.63, "percentage": 31.54, "elapsed_time": "2:40:26", "remaining_time": "5:48:14"}
+{"current_steps": 690, "total_steps": 2156, "loss": 1.2479, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.8393209176407223e-05, "epoch": 0.64, "percentage": 32.0, "elapsed_time": "2:42:45", "remaining_time": "5:45:48"}
+{"current_steps": 700, "total_steps": 2156, "loss": 1.2431, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.8084199592415305e-05, "epoch": 0.65, "percentage": 32.47, "elapsed_time": "2:45:05", "remaining_time": "5:43:23"}
+{"current_steps": 710, "total_steps": 2156, "loss": 1.245, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.777241194786591e-05, "epoch": 0.66, "percentage": 32.93, "elapsed_time": "2:47:32", "remaining_time": "5:41:14"}
+{"current_steps": 720, "total_steps": 2156, "loss": 1.2393, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.745791244206697e-05, "epoch": 0.67, "percentage": 33.4, "elapsed_time": "2:49:56", "remaining_time": "5:38:55"}
+{"current_steps": 730, "total_steps": 2156, "loss": 1.2473, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.714076785011359e-05, "epoch": 0.68, "percentage": 33.86, "elapsed_time": "2:52:20", "remaining_time": "5:36:38"}
+{"current_steps": 740, "total_steps": 2156, "loss": 1.2552, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.682104550871031e-05, "epoch": 0.69, "percentage": 34.32, "elapsed_time": "2:54:43", "remaining_time": "5:34:19"}
+{"current_steps": 750, "total_steps": 2156, "loss": 1.2356, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.649881330187401e-05, "epoch": 0.7, "percentage": 34.79, "elapsed_time": "2:57:03", "remaining_time": "5:31:56"}
+{"current_steps": 760, "total_steps": 2156, "loss": 1.2442, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.617413964652067e-05, "epoch": 0.7, "percentage": 35.25, "elapsed_time": "2:59:28", "remaining_time": "5:29:40"}
+{"current_steps": 770, "total_steps": 2156, "loss": 1.2605, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.5847093477938956e-05, "epoch": 0.71, "percentage": 35.71, "elapsed_time": "3:01:49", "remaining_time": "5:27:16"}
+{"current_steps": 780, "total_steps": 2156, "loss": 1.2364, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.551774423515378e-05, "epoch": 0.72, "percentage": 36.18, "elapsed_time": "3:04:10", "remaining_time": "5:24:55"}
+{"current_steps": 790, "total_steps": 2156, "loss": 1.2482, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.518616184618288e-05, "epoch": 0.73, "percentage": 36.64, "elapsed_time": "3:06:29", "remaining_time": "5:22:27"}
+{"current_steps": 800, "total_steps": 2156, "loss": 1.2327, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.4852416713189526e-05, "epoch": 0.74, "percentage": 37.11, "elapsed_time": "3:08:51", "remaining_time": "5:20:07"}
+{"current_steps": 810, "total_steps": 2156, "loss": 1.2548, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.4516579697534705e-05, "epoch": 0.75, "percentage": 37.57, "elapsed_time": "3:11:15", "remaining_time": "5:17:48"}
+{"current_steps": 820, "total_steps": 2156, "loss": 1.2469, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.417872210473162e-05, "epoch": 0.76, "percentage": 38.03, "elapsed_time": "3:13:34", "remaining_time": "5:15:22"}
+{"current_steps": 830, "total_steps": 2156, "loss": 1.2424, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.3838915669306034e-05, "epoch": 0.77, "percentage": 38.5, "elapsed_time": "3:15:54", "remaining_time": "5:12:59"}
+{"current_steps": 840, "total_steps": 2156, "loss": 1.2438, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.349723253956542e-05, "epoch": 0.78, "percentage": 38.96, "elapsed_time": "3:18:16", "remaining_time": "5:10:38"}
+{"current_steps": 850, "total_steps": 2156, "loss": 1.243, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.315374526228036e-05, "epoch": 0.79, "percentage": 39.42, "elapsed_time": "3:20:38", "remaining_time": "5:08:16"}
+{"current_steps": 860, "total_steps": 2156, "loss": 1.2416, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.2808526767281225e-05, "epoch": 0.8, "percentage": 39.89, "elapsed_time": "3:22:59", "remaining_time": "5:05:54"}
+{"current_steps": 870, "total_steps": 2156, "loss": 1.2399, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.246165035197364e-05, "epoch": 0.81, "percentage": 40.35, "elapsed_time": "3:25:22", "remaining_time": "5:03:34"}
+{"current_steps": 880, "total_steps": 2156, "loss": 1.239, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.211318966577581e-05, "epoch": 0.82, "percentage": 40.82, "elapsed_time": "3:27:41", "remaining_time": "5:01:08"}
+{"current_steps": 890, "total_steps": 2156, "loss": 1.2498, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.176321869448116e-05, "epoch": 0.83, "percentage": 41.28, "elapsed_time": "3:30:03", "remaining_time": "4:58:48"}
+{"current_steps": 900, "total_steps": 2156, "loss": 1.2438, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.1411811744549536e-05, "epoch": 0.83, "percentage": 41.74, "elapsed_time": "3:32:27", "remaining_time": "4:56:29"}
+{"current_steps": 910, "total_steps": 2156, "loss": 1.233, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.105904342733032e-05, "epoch": 0.84, "percentage": 42.21, "elapsed_time": "3:34:51", "remaining_time": "4:54:11"}
+{"current_steps": 920, "total_steps": 2156, "loss": 1.2315, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.070498864322081e-05, "epoch": 0.85, "percentage": 42.67, "elapsed_time": "3:37:13", "remaining_time": "4:51:50"}
+{"current_steps": 930, "total_steps": 2156, "loss": 1.232, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.034972256576328e-05, "epoch": 0.86, "percentage": 43.14, "elapsed_time": "3:39:35", "remaining_time": "4:49:28"}
+{"current_steps": 940, "total_steps": 2156, "loss": 1.2467, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.999332062568395e-05, "epoch": 0.87, "percentage": 43.6, "elapsed_time": "3:42:00", "remaining_time": "4:47:11"}
+{"current_steps": 950, "total_steps": 2156, "loss": 1.2483, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.9635858494877384e-05, "epoch": 0.88, "percentage": 44.06, "elapsed_time": "3:44:20", "remaining_time": "4:44:47"}
+{"current_steps": 960, "total_steps": 2156, "loss": 1.2377, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.9277412070339782e-05, "epoch": 0.89, "percentage": 44.53, "elapsed_time": "3:46:43", "remaining_time": "4:42:27"}
+{"current_steps": 970, "total_steps": 2156, "loss": 1.2347, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.891805745805429e-05, "epoch": 0.9, "percentage": 44.99, "elapsed_time": "3:49:02", "remaining_time": "4:40:03"}
+{"current_steps": 980, "total_steps": 2156, "loss": 1.2428, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.8557870956832132e-05, "epoch": 0.91, "percentage": 45.45, "elapsed_time": "3:51:26", "remaining_time": "4:37:44"}
+{"current_steps": 990, "total_steps": 2156, "loss": 1.244, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.8196929042112652e-05, "epoch": 0.92, "percentage": 45.92, "elapsed_time": "3:53:49", "remaining_time": "4:35:24"}
+{"current_steps": 1000, "total_steps": 2156, "loss": 1.2317, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.783530834972594e-05, "epoch": 0.93, "percentage": 46.38, "elapsed_time": "3:56:15", "remaining_time": "4:33:06"}
+{"current_steps": 1010, "total_steps": 2156, "loss": 1.2421, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.7473085659621377e-05, "epoch": 0.94, "percentage": 46.85, "elapsed_time": "3:58:37", "remaining_time": "4:30:45"}
+{"current_steps": 1020, "total_steps": 2156, "loss": 1.2446, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.711033787956555e-05, "epoch": 0.95, "percentage": 47.31, "elapsed_time": "4:01:01", "remaining_time": "4:28:26"}
+{"current_steps": 1030, "total_steps": 2156, "loss": 1.2471, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.6747142028813105e-05, "epoch": 0.96, "percentage": 47.77, "elapsed_time": "4:03:24", "remaining_time": "4:26:05"}
+{"current_steps": 1040, "total_steps": 2156, "loss": 1.2403, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.638357522175383e-05, "epoch": 0.96, "percentage": 48.24, "elapsed_time": "4:05:43", "remaining_time": "4:23:41"}
+{"current_steps": 1050, "total_steps": 2156, "loss": 1.2413, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.6019714651539646e-05, "epoch": 0.97, "percentage": 48.7, "elapsed_time": "4:08:11", "remaining_time": "4:21:25"}
+{"current_steps": 1060, "total_steps": 2156, "loss": 1.2341, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.565563757369475e-05, "epoch": 0.98, "percentage": 49.17, "elapsed_time": "4:10:31", "remaining_time": "4:19:02"}
+{"current_steps": 1070, "total_steps": 2156, "loss": 1.236, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.529142128971268e-05, "epoch": 0.99, "percentage": 49.63, "elapsed_time": "4:12:53", "remaining_time": "4:16:39"}
+{"current_steps": 1080, "total_steps": 2156, "loss": 1.2423, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.492714313064342e-05, "epoch": 1.0, "percentage": 50.09, "elapsed_time": "4:15:12", "remaining_time": "4:14:15"}
+{"current_steps": 1090, "total_steps": 2156, "loss": 1.2253, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.45628804406744e-05, "epoch": 1.01, "percentage": 50.56, "elapsed_time": "4:17:33", "remaining_time": "4:11:53"}
+{"current_steps": 1100, "total_steps": 2156, "loss": 1.2377, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.419871056070862e-05, "epoch": 1.02, "percentage": 51.02, "elapsed_time": "4:19:52", "remaining_time": "4:09:28"}
+{"current_steps": 1110, "total_steps": 2156, "loss": 1.2395, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.3834710811943514e-05, "epoch": 1.03, "percentage": 51.48, "elapsed_time": "4:22:18", "remaining_time": "4:07:11"}
+{"current_steps": 1120, "total_steps": 2156, "loss": 1.2238, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.3470958479453938e-05, "epoch": 1.04, "percentage": 51.95, "elapsed_time": "4:24:41", "remaining_time": "4:04:50"}
+{"current_steps": 1130, "total_steps": 2156, "loss": 1.2045, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.3107530795782877e-05, "epoch": 1.05, "percentage": 52.41, "elapsed_time": "4:27:04", "remaining_time": "4:02:29"}
+{"current_steps": 1140, "total_steps": 2156, "loss": 1.2302, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.2744504924543313e-05, "epoch": 1.06, "percentage": 52.88, "elapsed_time": "4:29:25", "remaining_time": "4:00:07"}
+{"current_steps": 1150, "total_steps": 2156, "loss": 1.233, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.23819579440347e-05, "epoch": 1.07, "percentage": 53.34, "elapsed_time": "4:31:52", "remaining_time": "3:57:49"}
+{"current_steps": 1160, "total_steps": 2156, "loss": 1.2341, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.2019966830877545e-05, "epoch": 1.08, "percentage": 53.8, "elapsed_time": "4:34:11", "remaining_time": "3:55:25"}
+{"current_steps": 1170, "total_steps": 2156, "loss": 1.2294, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.1658608443669635e-05, "epoch": 1.08, "percentage": 54.27, "elapsed_time": "4:36:35", "remaining_time": "3:53:05"}
+{"current_steps": 1180, "total_steps": 2156, "loss": 1.2308, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.1297959506667224e-05, "epoch": 1.09, "percentage": 54.73, "elapsed_time": "4:38:55", "remaining_time": "3:50:42"}
+{"current_steps": 1190, "total_steps": 2156, "loss": 1.2427, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.0938096593494855e-05, "epoch": 1.1, "percentage": 55.19, "elapsed_time": "4:41:12", "remaining_time": "3:48:16"}
+{"current_steps": 1200, "total_steps": 2156, "loss": 1.2404, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.057909611088709e-05, "epoch": 1.11, "percentage": 55.66, "elapsed_time": "4:43:32", "remaining_time": "3:45:53"}
+{"current_steps": 1210, "total_steps": 2156, "loss": 1.2245, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.02210342824657e-05, "epoch": 1.12, "percentage": 56.12, "elapsed_time": "4:45:58", "remaining_time": "3:43:35"}
+{"current_steps": 1220, "total_steps": 2156, "loss": 1.2288, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.9863987132555706e-05, "epoch": 1.13, "percentage": 56.59, "elapsed_time": "4:48:20", "remaining_time": "3:41:13"}
+{"current_steps": 1230, "total_steps": 2156, "loss": 1.2401, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.9508030470043806e-05, "epoch": 1.14, "percentage": 57.05, "elapsed_time": "4:50:46", "remaining_time": "3:38:54"}
+{"current_steps": 1240, "total_steps": 2156, "loss": 1.2275, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.915323987228247e-05, "epoch": 1.15, "percentage": 57.51, "elapsed_time": "4:53:04", "remaining_time": "3:36:29"}
+{"current_steps": 1250, "total_steps": 2156, "loss": 1.2251, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.8799690669043212e-05, "epoch": 1.16, "percentage": 57.98, "elapsed_time": "4:55:27", "remaining_time": "3:34:08"}
+{"current_steps": 1260, "total_steps": 2156, "loss": 1.238, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.8447457926522454e-05, "epoch": 1.17, "percentage": 58.44, "elapsed_time": "4:57:48", "remaining_time": "3:31:46"}
+{"current_steps": 1270, "total_steps": 2156, "loss": 1.2291, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.8096616431403325e-05, "epoch": 1.18, "percentage": 58.91, "elapsed_time": "5:00:05", "remaining_time": "3:29:21"}
+{"current_steps": 1280, "total_steps": 2156, "loss": 1.2162, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.7747240674976857e-05, "epoch": 1.19, "percentage": 59.37, "elapsed_time": "5:02:27", "remaining_time": "3:26:59"}
+{"current_steps": 1290, "total_steps": 2156, "loss": 1.2315, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.7399404837325796e-05, "epoch": 1.2, "percentage": 59.83, "elapsed_time": "5:04:51", "remaining_time": "3:24:39"}
+{"current_steps": 1300, "total_steps": 2156, "loss": 1.2383, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.7053182771574633e-05, "epoch": 1.21, "percentage": 60.3, "elapsed_time": "5:07:10", "remaining_time": "3:22:15"}
+{"current_steps": 1310, "total_steps": 2156, "loss": 1.2254, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.6708647988208887e-05, "epoch": 1.21, "percentage": 60.76, "elapsed_time": "5:09:34", "remaining_time": "3:19:55"}
+{"current_steps": 1320, "total_steps": 2156, "loss": 1.2388, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.6365873639467315e-05, "epoch": 1.22, "percentage": 61.22, "elapsed_time": "5:11:56", "remaining_time": "3:17:34"}
+{"current_steps": 1330, "total_steps": 2156, "loss": 1.2213, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.602493250381003e-05, "epoch": 1.23, "percentage": 61.69, "elapsed_time": "5:14:18", "remaining_time": "3:15:11"}
+{"current_steps": 1340, "total_steps": 2156, "loss": 1.2366, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.5685896970466123e-05, "epoch": 1.24, "percentage": 62.15, "elapsed_time": "5:16:36", "remaining_time": "3:12:47"}
+{"current_steps": 1350, "total_steps": 2156, "loss": 1.2265, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.534883902406375e-05, "epoch": 1.25, "percentage": 62.62, "elapsed_time": "5:19:01", "remaining_time": "3:10:27"}
+{"current_steps": 1360, "total_steps": 2156, "loss": 1.2094, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.5013830229346326e-05, "epoch": 1.26, "percentage": 63.08, "elapsed_time": "5:21:24", "remaining_time": "3:08:07"}
+{"current_steps": 1370, "total_steps": 2156, "loss": 1.2314, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.4680941715977722e-05, "epoch": 1.27, "percentage": 63.54, "elapsed_time": "5:23:49", "remaining_time": "3:05:46"}
+{"current_steps": 1380, "total_steps": 2156, "loss": 1.2254, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.4350244163439892e-05, "epoch": 1.28, "percentage": 64.01, "elapsed_time": "5:26:07", "remaining_time": "3:03:23"}
+{"current_steps": 1390, "total_steps": 2156, "loss": 1.2287, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.4021807786026108e-05, "epoch": 1.29, "percentage": 64.47, "elapsed_time": "5:28:29", "remaining_time": "3:01:01"}
+{"current_steps": 1400, "total_steps": 2156, "loss": 1.2235, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.3695702317932862e-05, "epoch": 1.3, "percentage": 64.94, "elapsed_time": "5:30:55", "remaining_time": "2:58:42"}
+{"current_steps": 1410, "total_steps": 2156, "loss": 1.2209, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.337199699845387e-05, "epoch": 1.31, "percentage": 65.4, "elapsed_time": "5:33:10", "remaining_time": "2:56:16"}
+{"current_steps": 1420, "total_steps": 2156, "loss": 1.2296, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.3050760557279023e-05, "epoch": 1.32, "percentage": 65.86, "elapsed_time": "5:35:29", "remaining_time": "2:53:53"}
+{"current_steps": 1430, "total_steps": 2156, "loss": 1.2363, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.2732061199901562e-05, "epoch": 1.33, "percentage": 66.33, "elapsed_time": "5:37:52", "remaining_time": "2:51:32"}
+{"current_steps": 1440, "total_steps": 2156, "loss": 1.2116, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.2415966593136547e-05, "epoch": 1.34, "percentage": 66.79, "elapsed_time": "5:40:16", "remaining_time": "2:49:11"}
+{"current_steps": 1450, "total_steps": 2156, "loss": 1.2299, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.2102543850753808e-05, "epoch": 1.34, "percentage": 67.25, "elapsed_time": "5:42:38", "remaining_time": "2:46:50"}
+{"current_steps": 1460, "total_steps": 2156, "loss": 1.2204, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.1791859519228138e-05, "epoch": 1.35, "percentage": 67.72, "elapsed_time": "5:45:03", "remaining_time": "2:44:29"}
+{"current_steps": 1470, "total_steps": 2156, "loss": 1.2271, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.148397956361007e-05, "epoch": 1.36, "percentage": 68.18, "elapsed_time": "5:47:21", "remaining_time": "2:42:05"}
+{"current_steps": 1480, "total_steps": 2156, "loss": 1.2327, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.1178969353520018e-05, "epoch": 1.37, "percentage": 68.65, "elapsed_time": "5:49:44", "remaining_time": "2:39:44"}
+{"current_steps": 1490, "total_steps": 2156, "loss": 1.2252, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.087689364926897e-05, "epoch": 1.38, "percentage": 69.11, "elapsed_time": "5:52:04", "remaining_time": "2:37:22"}
+{"current_steps": 1500, "total_steps": 2156, "loss": 1.2305, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.0577816588108378e-05, "epoch": 1.39, "percentage": 69.57, "elapsed_time": "5:54:28", "remaining_time": "2:35:01"}
+{"current_steps": 1510, "total_steps": 2156, "loss": 1.2179, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.0281801670612449e-05, "epoch": 1.4, "percentage": 70.04, "elapsed_time": "5:56:53", "remaining_time": "2:32:40"}
+{"current_steps": 1520, "total_steps": 2156, "loss": 1.229, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.988911747195603e-06, "epoch": 1.41, "percentage": 70.5, "elapsed_time": "5:59:13", "remaining_time": "2:30:18"}
+{"current_steps": 1530, "total_steps": 2156, "loss": 1.2192, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.699209004767953e-06, "epoch": 1.42, "percentage": 70.96, "elapsed_time": "6:01:36", "remaining_time": "2:27:56"}
+{"current_steps": 1540, "total_steps": 2156, "loss": 1.2322, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.412754953531663e-06, "epoch": 1.43, "percentage": 71.43, "elapsed_time": "6:03:59", "remaining_time": "2:25:35"}
+{"current_steps": 1550, "total_steps": 2156, "loss": 1.2193, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.129610413921005e-06, "epoch": 1.44, "percentage": 71.89, "elapsed_time": "6:06:22", "remaining_time": "2:23:14"}
+{"current_steps": 1560, "total_steps": 2156, "loss": 1.2149, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 8.849835503688847e-06, "epoch": 1.45, "percentage": 72.36, "elapsed_time": "6:08:49", "remaining_time": "2:20:54"}
+{"current_steps": 1570, "total_steps": 2156, "loss": 1.217, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 8.57348962514234e-06, "epoch": 1.46, "percentage": 72.82, "elapsed_time": "6:11:09", "remaining_time": "2:18:32"}
+{"current_steps": 1580, "total_steps": 2156, "loss": 1.2368, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 8.30063145253053e-06, "epoch": 1.46, "percentage": 73.28, "elapsed_time": "6:13:27", "remaining_time": "2:16:08"}
+{"current_steps": 1590, "total_steps": 2156, "loss": 1.2264, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 8.031318919586523e-06, "epoch": 1.47, "percentage": 73.75, "elapsed_time": "6:15:52", "remaining_time": "2:13:48"}
+{"current_steps": 1600, "total_steps": 2156, "loss": 1.2243, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 7.7656092072269e-06, "epoch": 1.48, "percentage": 74.21, "elapsed_time": "6:18:11", "remaining_time": "2:11:25"}
+{"current_steps": 1610, "total_steps": 2156, "loss": 1.2427, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 7.503558731410959e-06, "epoch": 1.49, "percentage": 74.68, "elapsed_time": "6:20:32", "remaining_time": "2:09:03"}
+{"current_steps": 1620, "total_steps": 2156, "loss": 1.2207, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 7.245223131162376e-06, "epoch": 1.5, "percentage": 75.14, "elapsed_time": "6:22:56", "remaining_time": "2:06:42"}
+{"current_steps": 1630, "total_steps": 2156, "loss": 1.2211, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.9906572567558285e-06, "epoch": 1.51, "percentage": 75.6, "elapsed_time": "6:25:19", "remaining_time": "2:04:20"}
+{"current_steps": 1640, "total_steps": 2156, "loss": 1.2343, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.739915158071106e-06, "epoch": 1.52, "percentage": 76.07, "elapsed_time": "6:27:35", "remaining_time": "2:01:57"}
+{"current_steps": 1650, "total_steps": 2156, "loss": 1.2348, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.493050073117116e-06, "epoch": 1.53, "percentage": 76.53, "elapsed_time": "6:29:57", "remaining_time": "1:59:35"}
+{"current_steps": 1660, "total_steps": 2156, "loss": 1.2189, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.250114416728298e-06, "epoch": 1.54, "percentage": 76.99, "elapsed_time": "6:32:14", "remaining_time": "1:57:12"}
+{"current_steps": 1670, "total_steps": 2156, "loss": 1.2207, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.011159769435823e-06, "epoch": 1.55, "percentage": 77.46, "elapsed_time": "6:34:30", "remaining_time": "1:54:48"}
+{"current_steps": 1680, "total_steps": 2156, "loss": 1.2315, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.776236866515947e-06, "epoch": 1.56, "percentage": 77.92, "elapsed_time": "6:36:48", "remaining_time": "1:52:25"}
+{"current_steps": 1690, "total_steps": 2156, "loss": 1.2334, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.545395587217763e-06, "epoch": 1.57, "percentage": 78.39, "elapsed_time": "6:39:05", "remaining_time": "1:50:02"}
+{"current_steps": 1700, "total_steps": 2156, "loss": 1.232, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.318684944172752e-06, "epoch": 1.58, "percentage": 78.85, "elapsed_time": "6:41:25", "remaining_time": "1:47:40"}
+{"current_steps": 1710, "total_steps": 2156, "loss": 1.2235, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.096153072988366e-06, "epoch": 1.59, "percentage": 79.31, "elapsed_time": "6:43:44", "remaining_time": "1:45:18"}
+{"current_steps": 1720, "total_steps": 2156, "loss": 1.2244, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.8778472220277874e-06, "epoch": 1.59, "percentage": 79.78, "elapsed_time": "6:46:02", "remaining_time": "1:42:55"}
+{"current_steps": 1730, "total_steps": 2156, "loss": 1.2197, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.6638137423780395e-06, "epoch": 1.6, "percentage": 80.24, "elapsed_time": "6:48:25", "remaining_time": "1:40:34"}
+{"current_steps": 1740, "total_steps": 2156, "loss": 1.2242, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.454098078008667e-06, "epoch": 1.61, "percentage": 80.71, "elapsed_time": "6:50:51", "remaining_time": "1:38:13"}
+{"current_steps": 1750, "total_steps": 2156, "loss": 1.2382, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.248744756122986e-06, "epoch": 1.62, "percentage": 81.17, "elapsed_time": "6:53:10", "remaining_time": "1:35:51"}
+{"current_steps": 1760, "total_steps": 2156, "loss": 1.2265, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.047797377703985e-06, "epoch": 1.63, "percentage": 81.63, "elapsed_time": "6:55:29", "remaining_time": "1:33:29"}
+{"current_steps": 1770, "total_steps": 2156, "loss": 1.2153, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.851298608256892e-06, "epoch": 1.64, "percentage": 82.1, "elapsed_time": "6:57:52", "remaining_time": "1:31:07"}
+{"current_steps": 1780, "total_steps": 2156, "loss": 1.2198, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.6592901687503566e-06, "epoch": 1.65, "percentage": 82.56, "elapsed_time": "7:00:12", "remaining_time": "1:28:45"}
+{"current_steps": 1790, "total_steps": 2156, "loss": 1.23, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.471812826758178e-06, "epoch": 1.66, "percentage": 83.02, "elapsed_time": "7:02:31", "remaining_time": "1:26:23"}
+{"current_steps": 1800, "total_steps": 2156, "loss": 1.2268, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.288906387803464e-06, "epoch": 1.67, "percentage": 83.49, "elapsed_time": "7:04:53", "remaining_time": "1:24:02"}
+{"current_steps": 1810, "total_steps": 2156, "loss": 1.2228, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.1106096869070483e-06, "epoch": 1.68, "percentage": 83.95, "elapsed_time": "7:07:14", "remaining_time": "1:21:40"}
+{"current_steps": 1820, "total_steps": 2156, "loss": 1.2288, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.9369605803419715e-06, "epoch": 1.69, "percentage": 84.42, "elapsed_time": "7:09:30", "remaining_time": "1:19:17"}
+{"current_steps": 1830, "total_steps": 2156, "loss": 1.228, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.767995937595766e-06, "epoch": 1.7, "percentage": 84.88, "elapsed_time": "7:11:49", "remaining_time": "1:16:55"}
+{"current_steps": 1840, "total_steps": 2156, "loss": 1.2178, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.6037516335422728e-06, "epoch": 1.71, "percentage": 85.34, "elapsed_time": "7:14:09", "remaining_time": "1:14:33"}
+{"current_steps": 1850, "total_steps": 2156, "loss": 1.2386, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.4442625408246074e-06, "epoch": 1.72, "percentage": 85.81, "elapsed_time": "7:16:28", "remaining_time": "1:12:11"}
+{"current_steps": 1860, "total_steps": 2156, "loss": 1.2274, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.289562522450947e-06, "epoch": 1.72, "percentage": 86.27, "elapsed_time": "7:18:53", "remaining_time": "1:09:50"}
+{"current_steps": 1870, "total_steps": 2156, "loss": 1.2341, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.1396844246046903e-06, "epoch": 1.73, "percentage": 86.73, "elapsed_time": "7:21:17", "remaining_time": "1:07:29"}
+{"current_steps": 1880, "total_steps": 2156, "loss": 1.2118, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.9946600696704592e-06, "epoch": 1.74, "percentage": 87.2, "elapsed_time": "7:23:46", "remaining_time": "1:05:08"}
+{"current_steps": 1890, "total_steps": 2156, "loss": 1.2219, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.8545202494775509e-06, "epoch": 1.75, "percentage": 87.66, "elapsed_time": "7:26:09", "remaining_time": "1:02:47"}
+{"current_steps": 1900, "total_steps": 2156, "loss": 1.2268, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.7192947187621434e-06, "epoch": 1.76, "percentage": 88.13, "elapsed_time": "7:28:32", "remaining_time": "1:00:26"}
+{"current_steps": 1910, "total_steps": 2156, "loss": 1.2183, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.5890121888497366e-06, "epoch": 1.77, "percentage": 88.59, "elapsed_time": "7:30:53", "remaining_time": "0:58:04"}
+{"current_steps": 1920, "total_steps": 2156, "loss": 1.2253, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.463700321559075e-06, "epoch": 1.78, "percentage": 89.05, "elapsed_time": "7:33:19", "remaining_time": "0:55:43"}
+{"current_steps": 1930, "total_steps": 2156, "loss": 1.2262, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.3433857233289714e-06, "epoch": 1.79, "percentage": 89.52, "elapsed_time": "7:35:40", "remaining_time": "0:53:21"}
+{"current_steps": 1940, "total_steps": 2156, "loss": 1.2232, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.2280939395691859e-06, "epoch": 1.8, "percentage": 89.98, "elapsed_time": "7:37:59", "remaining_time": "0:50:59"}
+{"current_steps": 1950, "total_steps": 2156, "loss": 1.2376, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.1178494492365465e-06, "epoch": 1.81, "percentage": 90.45, "elapsed_time": "7:40:30", "remaining_time": "0:48:38"}
+{"current_steps": 1960, "total_steps": 2156, "loss": 1.2287, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.0126756596375686e-06, "epoch": 1.82, "percentage": 90.91, "elapsed_time": "7:42:48", "remaining_time": "0:46:16"}
+{"current_steps": 1970, "total_steps": 2156, "loss": 1.2354, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.125949014585383e-07, "epoch": 1.83, "percentage": 91.37, "elapsed_time": "7:45:10", "remaining_time": "0:43:55"}
+{"current_steps": 1980, "total_steps": 2156, "loss": 1.224, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 8.176284240242638e-07, "epoch": 1.84, "percentage": 91.84, "elapsed_time": "7:47:30", "remaining_time": "0:41:33"}
+{"current_steps": 1990, "total_steps": 2156, "loss": 1.2214, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 7.277963907863478e-07, "epoch": 1.85, "percentage": 92.3, "elapsed_time": "7:49:52", "remaining_time": "0:39:11"}
+{"current_steps": 2000, "total_steps": 2156, "loss": 1.2214, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.431178750420513e-07, "epoch": 1.85, "percentage": 92.76, "elapsed_time": "7:52:13", "remaining_time": "0:36:50"}
+{"current_steps": 2010, "total_steps": 2156, "loss": 1.225, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.636108558846359e-07, "epoch": 1.86, "percentage": 93.23, "elapsed_time": "7:54:36", "remaining_time": "0:34:28"}
+{"current_steps": 2020, "total_steps": 2156, "loss": 1.2373, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.892922143859918e-07, "epoch": 1.87, "percentage": 93.69, "elapsed_time": "7:56:57", "remaining_time": "0:32:06"}
+{"current_steps": 2030, "total_steps": 2156, "loss": 1.2268, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.201777300124249e-07, "epoch": 1.88, "percentage": 94.16, "elapsed_time": "7:59:17", "remaining_time": "0:29:44"}
+{"current_steps": 2040, "total_steps": 2156, "loss": 1.2209, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.562820772743392e-07, "epoch": 1.89, "percentage": 94.62, "elapsed_time": "8:01:43", "remaining_time": "0:27:23"}
+{"current_steps": 2050, "total_steps": 2156, "loss": 1.2184, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.9761882261050143e-07, "epoch": 1.9, "percentage": 95.08, "elapsed_time": "8:04:08", "remaining_time": "0:25:02"}
+{"current_steps": 2060, "total_steps": 2156, "loss": 1.2265, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.4420042150761215e-07, "epoch": 1.91, "percentage": 95.55, "elapsed_time": "8:06:28", "remaining_time": "0:22:40"}
+{"current_steps": 2070, "total_steps": 2156, "loss": 1.227, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.9603821585572058e-07, "epoch": 1.92, "percentage": 96.01, "elapsed_time": "8:08:51", "remaining_time": "0:20:18"}
+{"current_steps": 2080, "total_steps": 2156, "loss": 1.2172, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.531424315400931e-07, "epoch": 1.93, "percentage": 96.47, "elapsed_time": "8:11:20", "remaining_time": "0:17:57"}
+{"current_steps": 2090, "total_steps": 2156, "loss": 1.2196, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.1552217627004425e-07, "epoch": 1.94, "percentage": 96.94, "elapsed_time": "8:13:39", "remaining_time": "0:15:35"}
+{"current_steps": 2100, "total_steps": 2156, "loss": 1.2381, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 8.318543764516961e-08, "epoch": 1.95, "percentage": 97.4, "elapsed_time": "8:15:59", "remaining_time": "0:13:13"}
+{"current_steps": 2110, "total_steps": 2156, "loss": 1.2288, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.613908145939428e-08, "epoch": 1.96, "percentage": 97.87, "elapsed_time": "8:18:23", "remaining_time": "0:10:51"}
+{"current_steps": 2120, "total_steps": 2156, "loss": 1.2387, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.438885024322769e-08, "epoch": 1.97, "percentage": 98.33, "elapsed_time": "8:20:45", "remaining_time": "0:08:30"}
+{"current_steps": 2130, "total_steps": 2156, "loss": 1.2357, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.7939362044494534e-08, "epoch": 1.97, "percentage": 98.79, "elapsed_time": "8:23:04", "remaining_time": "0:06:08"}
+{"current_steps": 2140, "total_steps": 2156, "loss": 1.2338, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.794109447824659e-09, "epoch": 1.98, "percentage": 99.26, "elapsed_time": "8:25:26", "remaining_time": "0:03:46"}
+{"current_steps": 2150, "total_steps": 2156, "loss": 1.2226, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.554588330934478e-10, "epoch": 1.99, "percentage": 99.72, "elapsed_time": "8:27:54", "remaining_time": "0:01:25"}
+{"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": "8:29:20", "remaining_time": "0:00:00"}

trainer_state.json

@@ -0,0 +1,1318 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.999072786277237,
+  "eval_steps": 500,
+  "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": 1.6199,
+      "step": 10
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.998938447446803e-05,
+      "loss": 1.5062,
+      "step": 20
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 4.997611718058365e-05,
+      "loss": 1.4138,
+      "step": 30
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 4.9957546913022665e-05,
+      "loss": 1.369,
+      "step": 40
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 4.993367761465736e-05,
+      "loss": 1.3408,
+      "step": 50
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.9904514353459654e-05,
+      "loss": 1.3321,
+      "step": 60
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 4.9870063321425105e-05,
+      "loss": 1.3251,
+      "step": 70
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 4.983033183325818e-05,
+      "loss": 1.3228,
+      "step": 80
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 4.97853283248192e-05,
+      "loss": 1.3111,
+      "step": 90
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 4.973506235133323e-05,
+      "loss": 1.3013,
+      "step": 100
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 4.967954458536126e-05,
+      "loss": 1.3004,
+      "step": 110
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 4.9618786814534226e-05,
+      "loss": 1.2959,
+      "step": 120
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 4.955280193905022e-05,
+      "loss": 1.2969,
+      "step": 130
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.948160396893553e-05,
+      "loss": 1.2879,
+      "step": 140
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 4.9405208021069946e-05,
+      "loss": 1.277,
+      "step": 150
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 4.9323630315977156e-05,
+      "loss": 1.283,
+      "step": 160
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.9236888174380784e-05,
+      "loss": 1.288,
+      "step": 170
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 4.91450000135268e-05,
+      "loss": 1.287,
+      "step": 180
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.9047985343273154e-05,
+      "loss": 1.2726,
+      "step": 190
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.894586476194739e-05,
+      "loss": 1.2808,
+      "step": 200
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.883865995197319e-05,
+      "loss": 1.2657,
+      "step": 210
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.8726393675266716e-05,
+      "loss": 1.275,
+      "step": 220
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 4.860908976840376e-05,
+      "loss": 1.2667,
+      "step": 230
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 4.848677313755872e-05,
+      "loss": 1.2715,
+      "step": 240
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 4.835946975321647e-05,
+      "loss": 1.273,
+      "step": 250
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 4.822720664465827e-05,
+      "loss": 1.2605,
+      "step": 260
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 4.809001189422287e-05,
+      "loss": 1.2766,
+      "step": 270
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 4.794791463134399e-05,
+      "loss": 1.262,
+      "step": 280
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 4.780094502636552e-05,
+      "loss": 1.255,
+      "step": 290
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 4.764913428413572e-05,
+      "loss": 1.2652,
+      "step": 300
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 4.7492514637381727e-05,
+      "loss": 1.2668,
+      "step": 310
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 4.733111933986583e-05,
+      "loss": 1.2621,
+      "step": 320
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 4.716498265932501e-05,
+      "loss": 1.257,
+      "step": 330
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 4.699413987019512e-05,
+      "loss": 1.2789,
+      "step": 340
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 4.681862724612141e-05,
+      "loss": 1.2634,
+      "step": 350
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 4.663848205225674e-05,
+      "loss": 1.2594,
+      "step": 360
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 4.645374253734949e-05,
+      "loss": 1.26,
+      "step": 370
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 4.626444792562244e-05,
+      "loss": 1.2514,
+      "step": 380
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 4.607063840844463e-05,
+      "loss": 1.2506,
+      "step": 390
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 4.587235513579791e-05,
+      "loss": 1.2648,
+      "step": 400
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 4.5669640207539786e-05,
+      "loss": 1.2511,
+      "step": 410
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 4.546253666446484e-05,
+      "loss": 1.2594,
+      "step": 420
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 4.525108847916614e-05,
+      "loss": 1.2608,
+      "step": 430
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 4.503534054669892e-05,
+      "loss": 1.2597,
+      "step": 440
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 4.481533867504841e-05,
+      "loss": 1.2609,
+      "step": 450
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 4.4591129575403765e-05,
+      "loss": 1.2505,
+      "step": 460
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 4.43627608522403e-05,
+      "loss": 1.2481,
+      "step": 470
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 4.4130280993211974e-05,
+      "loss": 1.2612,
+      "step": 480
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 4.389373935885646e-05,
+      "loss": 1.2504,
+      "step": 490
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 4.365318617211479e-05,
+      "loss": 1.2518,
+      "step": 500
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 4.340867250766794e-05,
+      "loss": 1.2458,
+      "step": 510
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 4.316025028109258e-05,
+      "loss": 1.2345,
+      "step": 520
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 4.2907972237838225e-05,
+      "loss": 1.2521,
+      "step": 530
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 4.2651891942028274e-05,
+      "loss": 1.2528,
+      "step": 540
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 4.239206376508717e-05,
+      "loss": 1.2462,
+      "step": 550
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 4.212854287419611e-05,
+      "loss": 1.2401,
+      "step": 560
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 4.1861385220579934e-05,
+      "loss": 1.2496,
+      "step": 570
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 4.1590647527627404e-05,
+      "loss": 1.2522,
+      "step": 580
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 4.131638727884762e-05,
+      "loss": 1.2377,
+      "step": 590
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 4.103866270566498e-05,
+      "loss": 1.2467,
+      "step": 600
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 4.075753277505544e-05,
+      "loss": 1.2421,
+      "step": 610
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 4.0473057177026484e-05,
+      "loss": 1.2455,
+      "step": 620
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 4.018529631194369e-05,
+      "loss": 1.2294,
+      "step": 630
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 3.989431127770635e-05,
+      "loss": 1.2509,
+      "step": 640
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 3.960016385677513e-05,
+      "loss": 1.2354,
+      "step": 650
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 3.9302916503054246e-05,
+      "loss": 1.2318,
+      "step": 660
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 3.9002632328631164e-05,
+      "loss": 1.2376,
+      "step": 670
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 3.8699375090376534e-05,
+      "loss": 1.2412,
+      "step": 680
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 3.8393209176407223e-05,
+      "loss": 1.2479,
+      "step": 690
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 3.8084199592415305e-05,
+      "loss": 1.2431,
+      "step": 700
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 3.777241194786591e-05,
+      "loss": 1.245,
+      "step": 710
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 3.745791244206697e-05,
+      "loss": 1.2393,
+      "step": 720
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 3.714076785011359e-05,
+      "loss": 1.2473,
+      "step": 730
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 3.682104550871031e-05,
+      "loss": 1.2552,
+      "step": 740
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 3.649881330187401e-05,
+      "loss": 1.2356,
+      "step": 750
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 3.617413964652067e-05,
+      "loss": 1.2442,
+      "step": 760
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 3.5847093477938956e-05,
+      "loss": 1.2605,
+      "step": 770
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 3.551774423515378e-05,
+      "loss": 1.2364,
+      "step": 780
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 3.518616184618288e-05,
+      "loss": 1.2482,
+      "step": 790
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 3.4852416713189526e-05,
+      "loss": 1.2327,
+      "step": 800
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 3.4516579697534705e-05,
+      "loss": 1.2548,
+      "step": 810
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 3.417872210473162e-05,
+      "loss": 1.2469,
+      "step": 820
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 3.3838915669306034e-05,
+      "loss": 1.2424,
+      "step": 830
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 3.349723253956542e-05,
+      "loss": 1.2438,
+      "step": 840
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 3.315374526228036e-05,
+      "loss": 1.243,
+      "step": 850
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 3.2808526767281225e-05,
+      "loss": 1.2416,
+      "step": 860
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 3.246165035197364e-05,
+      "loss": 1.2399,
+      "step": 870
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 3.211318966577581e-05,
+      "loss": 1.239,
+      "step": 880
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 3.176321869448116e-05,
+      "loss": 1.2498,
+      "step": 890
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 3.1411811744549536e-05,
+      "loss": 1.2438,
+      "step": 900
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 3.105904342733032e-05,
+      "loss": 1.233,
+      "step": 910
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 3.070498864322081e-05,
+      "loss": 1.2315,
+      "step": 920
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 3.034972256576328e-05,
+      "loss": 1.232,
+      "step": 930
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 2.999332062568395e-05,
+      "loss": 1.2467,
+      "step": 940
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 2.9635858494877384e-05,
+      "loss": 1.2483,
+      "step": 950
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 2.9277412070339782e-05,
+      "loss": 1.2377,
+      "step": 960
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 2.891805745805429e-05,
+      "loss": 1.2347,
+      "step": 970
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 2.8557870956832132e-05,
+      "loss": 1.2428,
+      "step": 980
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 2.8196929042112652e-05,
+      "loss": 1.244,
+      "step": 990
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 2.783530834972594e-05,
+      "loss": 1.2317,
+      "step": 1000
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 2.7473085659621377e-05,
+      "loss": 1.2421,
+      "step": 1010
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 2.711033787956555e-05,
+      "loss": 1.2446,
+      "step": 1020
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 2.6747142028813105e-05,
+      "loss": 1.2471,
+      "step": 1030
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 2.638357522175383e-05,
+      "loss": 1.2403,
+      "step": 1040
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 2.6019714651539646e-05,
+      "loss": 1.2413,
+      "step": 1050
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 2.565563757369475e-05,
+      "loss": 1.2341,
+      "step": 1060
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 2.529142128971268e-05,
+      "loss": 1.236,
+      "step": 1070
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.492714313064342e-05,
+      "loss": 1.2423,
+      "step": 1080
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 2.45628804406744e-05,
+      "loss": 1.2253,
+      "step": 1090
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 2.419871056070862e-05,
+      "loss": 1.2377,
+      "step": 1100
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 2.3834710811943514e-05,
+      "loss": 1.2395,
+      "step": 1110
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 2.3470958479453938e-05,
+      "loss": 1.2238,
+      "step": 1120
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 2.3107530795782877e-05,
+      "loss": 1.2045,
+      "step": 1130
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 2.2744504924543313e-05,
+      "loss": 1.2302,
+      "step": 1140
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 2.23819579440347e-05,
+      "loss": 1.233,
+      "step": 1150
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 2.2019966830877545e-05,
+      "loss": 1.2341,
+      "step": 1160
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 2.1658608443669635e-05,
+      "loss": 1.2294,
+      "step": 1170
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 2.1297959506667224e-05,
+      "loss": 1.2308,
+      "step": 1180
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 2.0938096593494855e-05,
+      "loss": 1.2427,
+      "step": 1190
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 2.057909611088709e-05,
+      "loss": 1.2404,
+      "step": 1200
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 2.02210342824657e-05,
+      "loss": 1.2245,
+      "step": 1210
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 1.9863987132555706e-05,
+      "loss": 1.2288,
+      "step": 1220
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 1.9508030470043806e-05,
+      "loss": 1.2401,
+      "step": 1230
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 1.915323987228247e-05,
+      "loss": 1.2275,
+      "step": 1240
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 1.8799690669043212e-05,
+      "loss": 1.2251,
+      "step": 1250
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 1.8447457926522454e-05,
+      "loss": 1.238,
+      "step": 1260
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 1.8096616431403325e-05,
+      "loss": 1.2291,
+      "step": 1270
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 1.7747240674976857e-05,
+      "loss": 1.2162,
+      "step": 1280
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 1.7399404837325796e-05,
+      "loss": 1.2315,
+      "step": 1290
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.7053182771574633e-05,
+      "loss": 1.2383,
+      "step": 1300
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 1.6708647988208887e-05,
+      "loss": 1.2254,
+      "step": 1310
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 1.6365873639467315e-05,
+      "loss": 1.2388,
+      "step": 1320
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 1.602493250381003e-05,
+      "loss": 1.2213,
+      "step": 1330
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 1.5685896970466123e-05,
+      "loss": 1.2366,
+      "step": 1340
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 1.534883902406375e-05,
+      "loss": 1.2265,
+      "step": 1350
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 1.5013830229346326e-05,
+      "loss": 1.2094,
+      "step": 1360
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 1.4680941715977722e-05,
+      "loss": 1.2314,
+      "step": 1370
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 1.4350244163439892e-05,
+      "loss": 1.2254,
+      "step": 1380
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 1.4021807786026108e-05,
+      "loss": 1.2287,
+      "step": 1390
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 1.3695702317932862e-05,
+      "loss": 1.2235,
+      "step": 1400
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 1.337199699845387e-05,
+      "loss": 1.2209,
+      "step": 1410
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 1.3050760557279023e-05,
+      "loss": 1.2296,
+      "step": 1420
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 1.2732061199901562e-05,
+      "loss": 1.2363,
+      "step": 1430
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.2415966593136547e-05,
+      "loss": 1.2116,
+      "step": 1440
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 1.2102543850753808e-05,
+      "loss": 1.2299,
+      "step": 1450
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 1.1791859519228138e-05,
+      "loss": 1.2204,
+      "step": 1460
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 1.148397956361007e-05,
+      "loss": 1.2271,
+      "step": 1470
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 1.1178969353520018e-05,
+      "loss": 1.2327,
+      "step": 1480
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 1.087689364926897e-05,
+      "loss": 1.2252,
+      "step": 1490
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 1.0577816588108378e-05,
+      "loss": 1.2305,
+      "step": 1500
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 1.0281801670612449e-05,
+      "loss": 1.2179,
+      "step": 1510
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 9.988911747195603e-06,
+      "loss": 1.229,
+      "step": 1520
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 9.699209004767953e-06,
+      "loss": 1.2192,
+      "step": 1530
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 9.412754953531663e-06,
+      "loss": 1.2322,
+      "step": 1540
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 9.129610413921005e-06,
+      "loss": 1.2193,
+      "step": 1550
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 8.849835503688847e-06,
+      "loss": 1.2149,
+      "step": 1560
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 8.57348962514234e-06,
+      "loss": 1.217,
+      "step": 1570
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 8.30063145253053e-06,
+      "loss": 1.2368,
+      "step": 1580
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 8.031318919586523e-06,
+      "loss": 1.2264,
+      "step": 1590
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 7.7656092072269e-06,
+      "loss": 1.2243,
+      "step": 1600
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 7.503558731410959e-06,
+      "loss": 1.2427,
+      "step": 1610
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 7.245223131162376e-06,
+      "loss": 1.2207,
+      "step": 1620
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 6.9906572567558285e-06,
+      "loss": 1.2211,
+      "step": 1630
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 6.739915158071106e-06,
+      "loss": 1.2343,
+      "step": 1640
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 6.493050073117116e-06,
+      "loss": 1.2348,
+      "step": 1650
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 6.250114416728298e-06,
+      "loss": 1.2189,
+      "step": 1660
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 6.011159769435823e-06,
+      "loss": 1.2207,
+      "step": 1670
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 5.776236866515947e-06,
+      "loss": 1.2315,
+      "step": 1680
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 5.545395587217763e-06,
+      "loss": 1.2334,
+      "step": 1690
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 5.318684944172752e-06,
+      "loss": 1.232,
+      "step": 1700
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 5.096153072988366e-06,
+      "loss": 1.2235,
+      "step": 1710
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 4.8778472220277874e-06,
+      "loss": 1.2244,
+      "step": 1720
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 4.6638137423780395e-06,
+      "loss": 1.2197,
+      "step": 1730
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 4.454098078008667e-06,
+      "loss": 1.2242,
+      "step": 1740
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 4.248744756122986e-06,
+      "loss": 1.2382,
+      "step": 1750
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 4.047797377703985e-06,
+      "loss": 1.2265,
+      "step": 1760
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 3.851298608256892e-06,
+      "loss": 1.2153,
+      "step": 1770
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 3.6592901687503566e-06,
+      "loss": 1.2198,
+      "step": 1780
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 3.471812826758178e-06,
+      "loss": 1.23,
+      "step": 1790
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 3.288906387803464e-06,
+      "loss": 1.2268,
+      "step": 1800
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 3.1106096869070483e-06,
+      "loss": 1.2228,
+      "step": 1810
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 2.9369605803419715e-06,
+      "loss": 1.2288,
+      "step": 1820
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 2.767995937595766e-06,
+      "loss": 1.228,
+      "step": 1830
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 2.6037516335422728e-06,
+      "loss": 1.2178,
+      "step": 1840
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 2.4442625408246074e-06,
+      "loss": 1.2386,
+      "step": 1850
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 2.289562522450947e-06,
+      "loss": 1.2274,
+      "step": 1860
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 2.1396844246046903e-06,
+      "loss": 1.2341,
+      "step": 1870
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 1.9946600696704592e-06,
+      "loss": 1.2118,
+      "step": 1880
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 1.8545202494775509e-06,
+      "loss": 1.2219,
+      "step": 1890
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 1.7192947187621434e-06,
+      "loss": 1.2268,
+      "step": 1900
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 1.5890121888497366e-06,
+      "loss": 1.2183,
+      "step": 1910
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 1.463700321559075e-06,
+      "loss": 1.2253,
+      "step": 1920
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 1.3433857233289714e-06,
+      "loss": 1.2262,
+      "step": 1930
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 1.2280939395691859e-06,
+      "loss": 1.2232,
+      "step": 1940
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 1.1178494492365465e-06,
+      "loss": 1.2376,
+      "step": 1950
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 1.0126756596375686e-06,
+      "loss": 1.2287,
+      "step": 1960
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 9.125949014585383e-07,
+      "loss": 1.2354,
+      "step": 1970
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 8.176284240242638e-07,
+      "loss": 1.224,
+      "step": 1980
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 7.277963907863478e-07,
+      "loss": 1.2214,
+      "step": 1990
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 6.431178750420513e-07,
+      "loss": 1.2214,
+      "step": 2000
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 5.636108558846359e-07,
+      "loss": 1.225,
+      "step": 2010
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 4.892922143859918e-07,
+      "loss": 1.2373,
+      "step": 2020
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 4.201777300124249e-07,
+      "loss": 1.2268,
+      "step": 2030
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 3.562820772743392e-07,
+      "loss": 1.2209,
+      "step": 2040
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 2.9761882261050143e-07,
+      "loss": 1.2184,
+      "step": 2050
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 2.4420042150761215e-07,
+      "loss": 1.2265,
+      "step": 2060
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 1.9603821585572058e-07,
+      "loss": 1.227,
+      "step": 2070
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 1.531424315400931e-07,
+      "loss": 1.2172,
+      "step": 2080
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 1.1552217627004425e-07,
+      "loss": 1.2196,
+      "step": 2090
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 8.318543764516961e-08,
+      "loss": 1.2381,
+      "step": 2100
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 5.613908145939428e-08,
+      "loss": 1.2288,
+      "step": 2110
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 3.438885024322769e-08,
+      "loss": 1.2387,
+      "step": 2120
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 1.7939362044494534e-08,
+      "loss": 1.2357,
+      "step": 2130
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 6.794109447824659e-09,
+      "loss": 1.2338,
+      "step": 2140
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 9.554588330934478e-10,
+      "loss": 1.2226,
+      "step": 2150
+    },
+    {
+      "epoch": 2.0,
+      "step": 2156,
+      "total_flos": 3.386848502305784e+18,
+      "train_loss": 1.2461530792912217,
+      "train_runtime": 30560.3748,
+      "train_samples_per_second": 4.517,
+      "train_steps_per_second": 0.071
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 2156,
+  "num_train_epochs": 2,
+  "save_steps": 1000,
+  "total_flos": 3.386848502305784e+18,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e1ca0e4372bd1720a48da585ce6946399a06d0c984dcfcf5a2b7104827c4603f
+size 4219