Upload folder using huggingface_hub

README.md

@@ -0,0 +1,59 @@
+---
+license: other
+library_name: peft
+tags:
+- llama-factory
+- lora
+- generated_from_trainer
+base_model: baichuan-inc/Baichuan-7B
+model-index:
+- name: train_2024-04-24-13-17-50
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# train_2024-04-24-13-17-50
+
+This model is a fine-tuned version of [baichuan-inc/Baichuan-7B](https://huggingface.co/baichuan-inc/Baichuan-7B) on the alpaca_gpt4_zh and the alpaca_zh datasets.
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 0.0002
+- train_batch_size: 2
+- eval_batch_size: 8
+- seed: 42
+- gradient_accumulation_steps: 8
+- total_train_batch_size: 16
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- num_epochs: 3.0
+- mixed_precision_training: Native AMP
+
+### Training results
+
+
+
+### Framework versions
+
+- PEFT 0.10.0
+- Transformers 4.37.2
+- Pytorch 2.1.2+cu121
+- Datasets 2.19.0
+- Tokenizers 0.15.2

adapter_config.json

@@ -0,0 +1,28 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "baichuan-inc/Baichuan-7B",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 16,
+  "lora_dropout": 0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 8,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "W_pack"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dc78daf24caf5389f7e184abb48d6d551b8b0237f4d38b40783a3879fe2e72af
+size 16785760

all_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 3.0,
+    "train_loss": 1.628477378845215,
+    "train_runtime": 1045.5144,
+    "train_samples_per_second": 5.739,
+    "train_steps_per_second": 0.359
+}

checkpoint-100/README.md

@@ -0,0 +1,202 @@
+---
+library_name: peft
+base_model: baichuan-inc/Baichuan-7B
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.10.0

checkpoint-100/adapter_config.json

@@ -0,0 +1,28 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "baichuan-inc/Baichuan-7B",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 16,
+  "lora_dropout": 0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 8,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "W_pack"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}

checkpoint-100/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:964dbe013935f9a4a0622cfc65c00eb8a1b12a09be0c11bbff411331d1481bb4
+size 16785760

checkpoint-100/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e46fb0086b5ce076a8ddcab4e3f5e18ae88af0044725aaf062561a4836205b48
+size 33608634

checkpoint-100/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1ff264f99d31b522cc7e2a4eac9d38606d0c58a34c0adc74d71e0ca8b371dc36
+size 14244

checkpoint-100/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4edd7cee06c7bbd4d1f7764b46a812cfc788b88695132d1b9ae0efb29eb9229c
+size 1064

checkpoint-100/special_tokens_map.json

@@ -0,0 +1,24 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "</s>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  }
+}

checkpoint-100/tokenization_baichuan.py

@@ -0,0 +1,251 @@
+# coding=utf-8
+# Copyright 2022 EleutherAI and the HuggingFace Inc. team. All rights reserved.
+#
+# This code is based on EleutherAI's GPT-NeoX library and the GPT-NeoX
+# and OPT implementations in this library. It has been modified from its
+# original forms to accommodate minor architectural differences compared
+# to GPT-NeoX and OPT used by the Meta AI team that trained the model.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import os
+from shutil import copyfile
+from typing import Any, Dict, List, Optional, Tuple
+
+import sentencepiece as spm
+
+from transformers.tokenization_utils import AddedToken, PreTrainedTokenizer
+from transformers.utils import logging
+
+
+logger = logging.get_logger(__name__)
+
+VOCAB_FILES_NAMES = {"vocab_file": "tokenizer.model"}
+
+PRETRAINED_VOCAB_FILES_MAP = {
+    "vocab_file": {},
+    "tokenizer_file": {},
+}
+PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES = {}
+
+
+class BaiChuanTokenizer(PreTrainedTokenizer):
+    """
+    Construct a BaiChuan tokenizer. Based on byte-level Byte-Pair-Encoding.
+
+    Args:
+        vocab_file (`str`):
+            Path to the vocabulary file.
+    """
+
+    vocab_files_names = VOCAB_FILES_NAMES
+    pretrained_vocab_files_map = PRETRAINED_VOCAB_FILES_MAP
+    max_model_input_sizes = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES
+    model_input_names = ["input_ids", "attention_mask"]
+
+    def __init__(
+        self,
+        vocab_file,
+        unk_token="<unk>",
+        bos_token="<s>",
+        eos_token="</s>",
+        pad_token=None,
+        sp_model_kwargs: Optional[Dict[str, Any]] = None,
+        add_bos_token=True,
+        add_eos_token=False,
+        clean_up_tokenization_spaces=False,
+        **kwargs,
+    ):
+        self.sp_model_kwargs = {} if sp_model_kwargs is None else sp_model_kwargs
+        bos_token = AddedToken(bos_token, lstrip=False, rstrip=False) if isinstance(bos_token, str) else bos_token
+        eos_token = AddedToken(eos_token, lstrip=False, rstrip=False) if isinstance(eos_token, str) else eos_token
+        unk_token = AddedToken(unk_token, lstrip=False, rstrip=False) if isinstance(unk_token, str) else unk_token
+        pad_token = AddedToken(pad_token, lstrip=False, rstrip=False) if isinstance(pad_token, str) else pad_token
+        self.vocab_file = vocab_file
+        self.add_bos_token = add_bos_token
+        self.add_eos_token = add_eos_token
+        self.sp_model = spm.SentencePieceProcessor(**self.sp_model_kwargs)
+        self.sp_model.Load(vocab_file)
+        
+        super().__init__(
+            bos_token=bos_token,
+            eos_token=eos_token,
+            unk_token=unk_token,
+            pad_token=pad_token,
+            add_bos_token=add_bos_token,
+            add_eos_token=add_eos_token,
+            sp_model_kwargs=self.sp_model_kwargs,
+            clean_up_tokenization_spaces=clean_up_tokenization_spaces,
+            **kwargs,
+        )
+
+    def __getstate__(self):
+        state = self.__dict__.copy()
+        state["sp_model"] = None
+        return state
+
+    def __setstate__(self, d):
+        self.__dict__ = d
+        self.sp_model = spm.SentencePieceProcessor(**self.sp_model_kwargs)
+        self.sp_model.Load(self.vocab_file)
+
+    @property
+    def vocab_size(self):
+        """Returns vocab size"""
+        return self.sp_model.get_piece_size()
+
+    def get_vocab(self):
+        """Returns vocab as a dict"""
+        vocab = {self.convert_ids_to_tokens(i): i for i in range(self.vocab_size)}
+        vocab.update(self.added_tokens_encoder)
+        return vocab
+
+    def _tokenize(self, text):
+        """Returns a tokenized string."""
+        return self.sp_model.encode(text, out_type=str)
+
+    def _convert_token_to_id(self, token):
+        """Converts a token (str) in an id using the vocab."""
+        return self.sp_model.piece_to_id(token)
+
+    def _convert_id_to_token(self, index):
+        """Converts an index (integer) in a token (str) using the vocab."""
+        token = self.sp_model.IdToPiece(index)
+        return token
+
+    def convert_tokens_to_string(self, tokens):
+        """Converts a sequence of tokens (string) in a single string."""
+        current_sub_tokens = []
+        out_string = ""
+        prev_is_special = False
+        for i, token in enumerate(tokens):
+            # make sure that special tokens are not decoded using sentencepiece model
+            if token in self.all_special_tokens:
+                if not prev_is_special and i != 0:
+                    out_string += " "
+                out_string += self.sp_model.decode(current_sub_tokens) + token
+                prev_is_special = True
+                current_sub_tokens = []
+            else:
+                current_sub_tokens.append(token)
+                prev_is_special = False
+        out_string += self.sp_model.decode(current_sub_tokens)
+        return out_string
+
+    def save_vocabulary(self, save_directory, filename_prefix: Optional[str] = None) -> Tuple[str]:
+        """
+        Save the vocabulary and special tokens file to a directory.
+
+        Args:
+            save_directory (`str`):
+                The directory in which to save the vocabulary.
+
+        Returns:
+            `Tuple(str)`: Paths to the files saved.
+        """
+        if not os.path.isdir(save_directory):
+            logger.error(f"Vocabulary path ({save_directory}) should be a directory")
+            return
+        out_vocab_file = os.path.join(
+            save_directory, (filename_prefix + "-" if filename_prefix else "") + VOCAB_FILES_NAMES["vocab_file"]
+        )
+
+        if os.path.abspath(self.vocab_file) != os.path.abspath(out_vocab_file) and os.path.isfile(self.vocab_file):
+            copyfile(self.vocab_file, out_vocab_file)
+        elif not os.path.isfile(self.vocab_file):
+            with open(out_vocab_file, "wb") as fi:
+                content_spiece_model = self.sp_model.serialized_model_proto()
+                fi.write(content_spiece_model)
+
+        return (out_vocab_file,)
+
+    def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=None):
+        bos_token_id = [self.bos_token_id] if self.add_bos_token else []
+        eos_token_id = [self.eos_token_id] if self.add_eos_token else []
+
+        output = bos_token_id + token_ids_0 + eos_token_id
+
+        if token_ids_1 is not None:
+            output = output + bos_token_id + token_ids_1 + eos_token_id
+
+        return output
+
+    def get_special_tokens_mask(
+        self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None, already_has_special_tokens: bool = False
+    ) -> List[int]:
+        """
+        Retrieve sequence ids from a token list that has no special tokens added. This method is called when adding
+        special tokens using the tokenizer `prepare_for_model` method.
+
+        Args:
+            token_ids_0 (`List[int]`):
+                List of IDs.
+            token_ids_1 (`List[int]`, *optional*):
+                Optional second list of IDs for sequence pairs.
+            already_has_special_tokens (`bool`, *optional*, defaults to `False`):
+                Whether or not the token list is already formatted with special tokens for the model.
+
+        Returns:
+            `List[int]`: A list of integers in the range [0, 1]: 1 for a special token, 0 for a sequence token.
+        """
+        if already_has_special_tokens:
+            return super().get_special_tokens_mask(
+                token_ids_0=token_ids_0, token_ids_1=token_ids_1, already_has_special_tokens=True
+            )
+
+        bos_token_id = [1] if self.add_bos_token else []
+        eos_token_id = [1] if self.add_eos_token else []
+
+        if token_ids_1 is None:
+            return bos_token_id + ([0] * len(token_ids_0)) + eos_token_id
+        return (
+            bos_token_id
+            + ([0] * len(token_ids_0))
+            + eos_token_id
+            + bos_token_id
+            + ([0] * len(token_ids_1))
+            + eos_token_id
+        )
+
+    def create_token_type_ids_from_sequences(
+        self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None
+    ) -> List[int]:
+        """
+        Creates a mask from the two sequences passed to be used in a sequence-pair classification task. An ALBERT
+        sequence pair mask has the following format:
+
+        ```
+        0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1
+        | first sequence    | second sequence |
+        ```
+
+        if token_ids_1 is None, only returns the first portion of the mask (0s).
+
+        Args:
+            token_ids_0 (`List[int]`):
+                List of ids.
+            token_ids_1 (`List[int]`, *optional*):
+                Optional second list of IDs for sequence pairs.
+
+        Returns:
+            `List[int]`: List of [token type IDs](../glossary#token-type-ids) according to the given sequence(s).
+        """
+        bos_token_id = [self.bos_token_id] if self.add_bos_token else []
+        eos_token_id = [self.eos_token_id] if self.add_eos_token else []
+
+        output = [0] * len(bos_token_id + token_ids_0 + eos_token_id)
+
+        if token_ids_1 is not None:
+            output += [1] * len(bos_token_id + token_ids_1 + eos_token_id)
+
+        return output

checkpoint-100/tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4be54af290d93c113bcbf421115ae9eed9d6340408f564898f1e966dc738ef01
+size 1136699

checkpoint-100/tokenizer_config.json

@@ -0,0 +1,47 @@
+{
+  "add_bos_token": false,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "auto_map": {
+    "AutoTokenizer": [
+      "tokenization_baichuan.BaiChuanTokenizer",
+      null
+    ]
+  },
+  "bos_token": "<s>",
+  "chat_template": "{% if messages[0]['role'] == 'system' %}{% set system_message = messages[0]['content'] %}{% endif %}{% if system_message is defined %}{{ system_message + '\\n' }}{% endif %}{% for message in messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ 'Human: ' + content + '\\nAssistant: ' }}{% elif message['role'] == 'assistant' %}{{ content + '</s>' + '\\n' }}{% endif %}{% endfor %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "</s>",
+  "padding_side": "right",
+  "sp_model_kwargs": {},
+  "split_special_tokens": false,
+  "tokenizer_class": "BaiChuanTokenizer",
+  "unk_token": "<unk>"
+}

checkpoint-100/trainer_state.json

@@ -0,0 +1,141 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.8,
+  "eval_steps": 500,
+  "global_step": 100,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.04,
+      "learning_rate": 0.00019991228300988585,
+      "loss": 2.1829,
+      "step": 5
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 0.00019964928592495045,
+      "loss": 1.8983,
+      "step": 10
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 0.0001992114701314478,
+      "loss": 1.7871,
+      "step": 15
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 0.0001985996037070505,
+      "loss": 1.8482,
+      "step": 20
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 0.00019781476007338058,
+      "loss": 1.731,
+      "step": 25
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 0.0001968583161128631,
+      "loss": 1.6797,
+      "step": 30
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 0.00019573194975320673,
+      "loss": 1.6694,
+      "step": 35
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 0.00019443763702374812,
+      "loss": 1.7493,
+      "step": 40
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 0.00019297764858882514,
+      "loss": 1.7823,
+      "step": 45
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 0.0001913545457642601,
+      "loss": 1.6518,
+      "step": 50
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 0.0001895711760239413,
+      "loss": 1.7334,
+      "step": 55
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 0.00018763066800438636,
+      "loss": 1.7565,
+      "step": 60
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 0.00018553642601605068,
+      "loss": 1.6493,
+      "step": 65
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 0.00018329212407100994,
+      "loss": 1.7431,
+      "step": 70
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 0.00018090169943749476,
+      "loss": 1.7561,
+      "step": 75
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 0.000178369345732584,
+      "loss": 1.6185,
+      "step": 80
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 0.00017569950556517566,
+      "loss": 1.6059,
+      "step": 85
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 0.00017289686274214118,
+      "loss": 1.772,
+      "step": 90
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 0.00016996633405133655,
+      "loss": 1.5847,
+      "step": 95
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 0.00016691306063588583,
+      "loss": 1.615,
+      "step": 100
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 375,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 100,
+  "total_flos": 1.167000136777728e+16,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-100/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:84990e28d454e290f6393201e221ba051e01af18e581a4f5994ac8396ad7c48b
+size 4920

checkpoint-200/README.md

@@ -0,0 +1,202 @@
+---
+library_name: peft
+base_model: baichuan-inc/Baichuan-7B
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.10.0

checkpoint-200/adapter_config.json

@@ -0,0 +1,28 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "baichuan-inc/Baichuan-7B",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 16,
+  "lora_dropout": 0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 8,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "W_pack"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}

checkpoint-200/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9fd427937ffad205d951d665d21e6c29ec78ea7ca2c0874c99af7bce99bf6891
+size 16785760

checkpoint-200/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b191c5fd1b3719f54e10d217585d462494ad34105326a4ade27909c4ade32a32
+size 33608634

checkpoint-200/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8d138cfe3a4adf21f048848ee35837c9a757a0a3616ff7adbb45b69aac247435
+size 14244

checkpoint-200/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:113d3d57baf5ea6fa1f49e5c9ccd10b9f9f78d00022910057b72b5bcea1f1d14
+size 1064

checkpoint-200/special_tokens_map.json

@@ -0,0 +1,24 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "</s>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  }
+}

checkpoint-200/tokenization_baichuan.py

@@ -0,0 +1,251 @@
+# coding=utf-8
+# Copyright 2022 EleutherAI and the HuggingFace Inc. team. All rights reserved.
+#
+# This code is based on EleutherAI's GPT-NeoX library and the GPT-NeoX
+# and OPT implementations in this library. It has been modified from its
+# original forms to accommodate minor architectural differences compared
+# to GPT-NeoX and OPT used by the Meta AI team that trained the model.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import os
+from shutil import copyfile
+from typing import Any, Dict, List, Optional, Tuple
+
+import sentencepiece as spm
+
+from transformers.tokenization_utils import AddedToken, PreTrainedTokenizer
+from transformers.utils import logging
+
+
+logger = logging.get_logger(__name__)
+
+VOCAB_FILES_NAMES = {"vocab_file": "tokenizer.model"}
+
+PRETRAINED_VOCAB_FILES_MAP = {
+    "vocab_file": {},
+    "tokenizer_file": {},
+}
+PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES = {}
+
+
+class BaiChuanTokenizer(PreTrainedTokenizer):
+    """
+    Construct a BaiChuan tokenizer. Based on byte-level Byte-Pair-Encoding.
+
+    Args:
+        vocab_file (`str`):
+            Path to the vocabulary file.
+    """
+
+    vocab_files_names = VOCAB_FILES_NAMES
+    pretrained_vocab_files_map = PRETRAINED_VOCAB_FILES_MAP
+    max_model_input_sizes = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES
+    model_input_names = ["input_ids", "attention_mask"]
+
+    def __init__(
+        self,
+        vocab_file,
+        unk_token="<unk>",
+        bos_token="<s>",
+        eos_token="</s>",
+        pad_token=None,
+        sp_model_kwargs: Optional[Dict[str, Any]] = None,
+        add_bos_token=True,
+        add_eos_token=False,
+        clean_up_tokenization_spaces=False,
+        **kwargs,
+    ):
+        self.sp_model_kwargs = {} if sp_model_kwargs is None else sp_model_kwargs
+        bos_token = AddedToken(bos_token, lstrip=False, rstrip=False) if isinstance(bos_token, str) else bos_token
+        eos_token = AddedToken(eos_token, lstrip=False, rstrip=False) if isinstance(eos_token, str) else eos_token
+        unk_token = AddedToken(unk_token, lstrip=False, rstrip=False) if isinstance(unk_token, str) else unk_token
+        pad_token = AddedToken(pad_token, lstrip=False, rstrip=False) if isinstance(pad_token, str) else pad_token
+        self.vocab_file = vocab_file
+        self.add_bos_token = add_bos_token
+        self.add_eos_token = add_eos_token
+        self.sp_model = spm.SentencePieceProcessor(**self.sp_model_kwargs)
+        self.sp_model.Load(vocab_file)
+        
+        super().__init__(
+            bos_token=bos_token,
+            eos_token=eos_token,
+            unk_token=unk_token,
+            pad_token=pad_token,
+            add_bos_token=add_bos_token,
+            add_eos_token=add_eos_token,
+            sp_model_kwargs=self.sp_model_kwargs,
+            clean_up_tokenization_spaces=clean_up_tokenization_spaces,
+            **kwargs,
+        )
+
+    def __getstate__(self):
+        state = self.__dict__.copy()
+        state["sp_model"] = None
+        return state
+
+    def __setstate__(self, d):
+        self.__dict__ = d
+        self.sp_model = spm.SentencePieceProcessor(**self.sp_model_kwargs)
+        self.sp_model.Load(self.vocab_file)
+
+    @property
+    def vocab_size(self):
+        """Returns vocab size"""
+        return self.sp_model.get_piece_size()
+
+    def get_vocab(self):
+        """Returns vocab as a dict"""
+        vocab = {self.convert_ids_to_tokens(i): i for i in range(self.vocab_size)}
+        vocab.update(self.added_tokens_encoder)
+        return vocab
+
+    def _tokenize(self, text):
+        """Returns a tokenized string."""
+        return self.sp_model.encode(text, out_type=str)
+
+    def _convert_token_to_id(self, token):
+        """Converts a token (str) in an id using the vocab."""
+        return self.sp_model.piece_to_id(token)
+
+    def _convert_id_to_token(self, index):
+        """Converts an index (integer) in a token (str) using the vocab."""
+        token = self.sp_model.IdToPiece(index)
+        return token
+
+    def convert_tokens_to_string(self, tokens):
+        """Converts a sequence of tokens (string) in a single string."""
+        current_sub_tokens = []
+        out_string = ""
+        prev_is_special = False
+        for i, token in enumerate(tokens):
+            # make sure that special tokens are not decoded using sentencepiece model
+            if token in self.all_special_tokens:
+                if not prev_is_special and i != 0:
+                    out_string += " "
+                out_string += self.sp_model.decode(current_sub_tokens) + token
+                prev_is_special = True
+                current_sub_tokens = []
+            else:
+                current_sub_tokens.append(token)
+                prev_is_special = False
+        out_string += self.sp_model.decode(current_sub_tokens)
+        return out_string
+
+    def save_vocabulary(self, save_directory, filename_prefix: Optional[str] = None) -> Tuple[str]:
+        """
+        Save the vocabulary and special tokens file to a directory.
+
+        Args:
+            save_directory (`str`):
+                The directory in which to save the vocabulary.
+
+        Returns:
+            `Tuple(str)`: Paths to the files saved.
+        """
+        if not os.path.isdir(save_directory):
+            logger.error(f"Vocabulary path ({save_directory}) should be a directory")
+            return
+        out_vocab_file = os.path.join(
+            save_directory, (filename_prefix + "-" if filename_prefix else "") + VOCAB_FILES_NAMES["vocab_file"]
+        )
+
+        if os.path.abspath(self.vocab_file) != os.path.abspath(out_vocab_file) and os.path.isfile(self.vocab_file):
+            copyfile(self.vocab_file, out_vocab_file)
+        elif not os.path.isfile(self.vocab_file):
+            with open(out_vocab_file, "wb") as fi:
+                content_spiece_model = self.sp_model.serialized_model_proto()
+                fi.write(content_spiece_model)
+
+        return (out_vocab_file,)
+
+    def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=None):
+        bos_token_id = [self.bos_token_id] if self.add_bos_token else []
+        eos_token_id = [self.eos_token_id] if self.add_eos_token else []
+
+        output = bos_token_id + token_ids_0 + eos_token_id
+
+        if token_ids_1 is not None:
+            output = output + bos_token_id + token_ids_1 + eos_token_id
+
+        return output
+
+    def get_special_tokens_mask(
+        self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None, already_has_special_tokens: bool = False
+    ) -> List[int]:
+        """
+        Retrieve sequence ids from a token list that has no special tokens added. This method is called when adding
+        special tokens using the tokenizer `prepare_for_model` method.
+
+        Args:
+            token_ids_0 (`List[int]`):
+                List of IDs.
+            token_ids_1 (`List[int]`, *optional*):
+                Optional second list of IDs for sequence pairs.
+            already_has_special_tokens (`bool`, *optional*, defaults to `False`):
+                Whether or not the token list is already formatted with special tokens for the model.
+
+        Returns:
+            `List[int]`: A list of integers in the range [0, 1]: 1 for a special token, 0 for a sequence token.
+        """
+        if already_has_special_tokens:
+            return super().get_special_tokens_mask(
+                token_ids_0=token_ids_0, token_ids_1=token_ids_1, already_has_special_tokens=True
+            )
+
+        bos_token_id = [1] if self.add_bos_token else []
+        eos_token_id = [1] if self.add_eos_token else []
+
+        if token_ids_1 is None:
+            return bos_token_id + ([0] * len(token_ids_0)) + eos_token_id
+        return (
+            bos_token_id
+            + ([0] * len(token_ids_0))
+            + eos_token_id
+            + bos_token_id
+            + ([0] * len(token_ids_1))
+            + eos_token_id
+        )
+
+    def create_token_type_ids_from_sequences(
+        self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None
+    ) -> List[int]:
+        """
+        Creates a mask from the two sequences passed to be used in a sequence-pair classification task. An ALBERT
+        sequence pair mask has the following format:
+
+        ```
+        0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1
+        | first sequence    | second sequence |
+        ```
+
+        if token_ids_1 is None, only returns the first portion of the mask (0s).
+
+        Args:
+            token_ids_0 (`List[int]`):
+                List of ids.
+            token_ids_1 (`List[int]`, *optional*):
+                Optional second list of IDs for sequence pairs.
+
+        Returns:
+            `List[int]`: List of [token type IDs](../glossary#token-type-ids) according to the given sequence(s).
+        """
+        bos_token_id = [self.bos_token_id] if self.add_bos_token else []
+        eos_token_id = [self.eos_token_id] if self.add_eos_token else []
+
+        output = [0] * len(bos_token_id + token_ids_0 + eos_token_id)
+
+        if token_ids_1 is not None:
+            output += [1] * len(bos_token_id + token_ids_1 + eos_token_id)
+
+        return output

checkpoint-200/tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4be54af290d93c113bcbf421115ae9eed9d6340408f564898f1e966dc738ef01
+size 1136699

checkpoint-200/tokenizer_config.json

@@ -0,0 +1,47 @@
+{
+  "add_bos_token": false,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "auto_map": {
+    "AutoTokenizer": [
+      "tokenization_baichuan.BaiChuanTokenizer",
+      null
+    ]
+  },
+  "bos_token": "<s>",
+  "chat_template": "{% if messages[0]['role'] == 'system' %}{% set system_message = messages[0]['content'] %}{% endif %}{% if system_message is defined %}{{ system_message + '\\n' }}{% endif %}{% for message in messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ 'Human: ' + content + '\\nAssistant: ' }}{% elif message['role'] == 'assistant' %}{{ content + '</s>' + '\\n' }}{% endif %}{% endfor %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "</s>",
+  "padding_side": "right",
+  "sp_model_kwargs": {},
+  "split_special_tokens": false,
+  "tokenizer_class": "BaiChuanTokenizer",
+  "unk_token": "<unk>"
+}

checkpoint-200/trainer_state.json

@@ -0,0 +1,261 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.6,
+  "eval_steps": 500,
+  "global_step": 200,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.04,
+      "learning_rate": 0.00019991228300988585,
+      "loss": 2.1829,
+      "step": 5
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 0.00019964928592495045,
+      "loss": 1.8983,
+      "step": 10
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 0.0001992114701314478,
+      "loss": 1.7871,
+      "step": 15
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 0.0001985996037070505,
+      "loss": 1.8482,
+      "step": 20
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 0.00019781476007338058,
+      "loss": 1.731,
+      "step": 25
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 0.0001968583161128631,
+      "loss": 1.6797,
+      "step": 30
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 0.00019573194975320673,
+      "loss": 1.6694,
+      "step": 35
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 0.00019443763702374812,
+      "loss": 1.7493,
+      "step": 40
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 0.00019297764858882514,
+      "loss": 1.7823,
+      "step": 45
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 0.0001913545457642601,
+      "loss": 1.6518,
+      "step": 50
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 0.0001895711760239413,
+      "loss": 1.7334,
+      "step": 55
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 0.00018763066800438636,
+      "loss": 1.7565,
+      "step": 60
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 0.00018553642601605068,
+      "loss": 1.6493,
+      "step": 65
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 0.00018329212407100994,
+      "loss": 1.7431,
+      "step": 70
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 0.00018090169943749476,
+      "loss": 1.7561,
+      "step": 75
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 0.000178369345732584,
+      "loss": 1.6185,
+      "step": 80
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 0.00017569950556517566,
+      "loss": 1.6059,
+      "step": 85
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 0.00017289686274214118,
+      "loss": 1.772,
+      "step": 90
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 0.00016996633405133655,
+      "loss": 1.5847,
+      "step": 95
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 0.00016691306063588583,
+      "loss": 1.615,
+      "step": 100
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 0.000163742398974869,
+      "loss": 1.5706,
+      "step": 105
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 0.0001604599114862375,
+      "loss": 1.6997,
+      "step": 110
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 0.0001570713567684432,
+      "loss": 1.5529,
+      "step": 115
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 0.00015358267949789966,
+      "loss": 1.6631,
+      "step": 120
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 0.00015000000000000001,
+      "loss": 1.7483,
+      "step": 125
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 0.00014632960351198618,
+      "loss": 1.708,
+      "step": 130
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 0.00014257792915650728,
+      "loss": 1.6979,
+      "step": 135
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 0.0001387515586452103,
+      "loss": 1.53,
+      "step": 140
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 0.00013485720473218154,
+      "loss": 1.6821,
+      "step": 145
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 0.00013090169943749476,
+      "loss": 1.7208,
+      "step": 150
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 0.00012689198206152657,
+      "loss": 1.6841,
+      "step": 155
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 0.00012283508701106557,
+      "loss": 1.544,
+      "step": 160
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 0.00011873813145857249,
+      "loss": 1.5851,
+      "step": 165
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 0.00011460830285624118,
+      "loss": 1.56,
+      "step": 170
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 0.00011045284632676536,
+      "loss": 1.5691,
+      "step": 175
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 0.00010627905195293135,
+      "loss": 1.5201,
+      "step": 180
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 0.0001020942419883357,
+      "loss": 1.5098,
+      "step": 185
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 9.790575801166432e-05,
+      "loss": 1.5805,
+      "step": 190
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 9.372094804706867e-05,
+      "loss": 1.6742,
+      "step": 195
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 8.954715367323468e-05,
+      "loss": 1.5656,
+      "step": 200
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 375,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 100,
+  "total_flos": 2.33056963362816e+16,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-200/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:84990e28d454e290f6393201e221ba051e01af18e581a4f5994ac8396ad7c48b
+size 4920

checkpoint-300/README.md

@@ -0,0 +1,202 @@
+---
+library_name: peft
+base_model: baichuan-inc/Baichuan-7B
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.10.0

checkpoint-300/adapter_config.json

@@ -0,0 +1,28 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "baichuan-inc/Baichuan-7B",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 16,
+  "lora_dropout": 0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 8,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "W_pack"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}

checkpoint-300/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2307fb868a4009a4fc5a667539f2d876a9f07e72b7c1b4c860db994c35619fdc
+size 16785760

checkpoint-300/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2133b8491938d73143f5774b98ff887cce068cde7146f9669879447ceb31015f
+size 33608634

checkpoint-300/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c062f7f375beded48b5337f5a3f3a5cb38807fa3e85dbf3e294c0ab6b627bfc2
+size 14244

checkpoint-300/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:91b317c7ccd56c68fd734532b70e29880e29310f64235ace1ff27743b77de2c0
+size 1064

checkpoint-300/special_tokens_map.json

@@ -0,0 +1,24 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "</s>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  }
+}

checkpoint-300/tokenization_baichuan.py

@@ -0,0 +1,251 @@
+# coding=utf-8
+# Copyright 2022 EleutherAI and the HuggingFace Inc. team. All rights reserved.
+#
+# This code is based on EleutherAI's GPT-NeoX library and the GPT-NeoX
+# and OPT implementations in this library. It has been modified from its
+# original forms to accommodate minor architectural differences compared
+# to GPT-NeoX and OPT used by the Meta AI team that trained the model.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import os
+from shutil import copyfile
+from typing import Any, Dict, List, Optional, Tuple
+
+import sentencepiece as spm
+
+from transformers.tokenization_utils import AddedToken, PreTrainedTokenizer
+from transformers.utils import logging
+
+
+logger = logging.get_logger(__name__)
+
+VOCAB_FILES_NAMES = {"vocab_file": "tokenizer.model"}
+
+PRETRAINED_VOCAB_FILES_MAP = {
+    "vocab_file": {},
+    "tokenizer_file": {},
+}
+PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES = {}
+
+
+class BaiChuanTokenizer(PreTrainedTokenizer):
+    """
+    Construct a BaiChuan tokenizer. Based on byte-level Byte-Pair-Encoding.
+
+    Args:
+        vocab_file (`str`):
+            Path to the vocabulary file.
+    """
+
+    vocab_files_names = VOCAB_FILES_NAMES
+    pretrained_vocab_files_map = PRETRAINED_VOCAB_FILES_MAP
+    max_model_input_sizes = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES
+    model_input_names = ["input_ids", "attention_mask"]
+
+    def __init__(
+        self,
+        vocab_file,
+        unk_token="<unk>",
+        bos_token="<s>",
+        eos_token="</s>",
+        pad_token=None,
+        sp_model_kwargs: Optional[Dict[str, Any]] = None,
+        add_bos_token=True,
+        add_eos_token=False,
+        clean_up_tokenization_spaces=False,
+        **kwargs,
+    ):
+        self.sp_model_kwargs = {} if sp_model_kwargs is None else sp_model_kwargs
+        bos_token = AddedToken(bos_token, lstrip=False, rstrip=False) if isinstance(bos_token, str) else bos_token
+        eos_token = AddedToken(eos_token, lstrip=False, rstrip=False) if isinstance(eos_token, str) else eos_token
+        unk_token = AddedToken(unk_token, lstrip=False, rstrip=False) if isinstance(unk_token, str) else unk_token
+        pad_token = AddedToken(pad_token, lstrip=False, rstrip=False) if isinstance(pad_token, str) else pad_token
+        self.vocab_file = vocab_file
+        self.add_bos_token = add_bos_token
+        self.add_eos_token = add_eos_token
+        self.sp_model = spm.SentencePieceProcessor(**self.sp_model_kwargs)
+        self.sp_model.Load(vocab_file)
+        
+        super().__init__(
+            bos_token=bos_token,
+            eos_token=eos_token,
+            unk_token=unk_token,
+            pad_token=pad_token,
+            add_bos_token=add_bos_token,
+            add_eos_token=add_eos_token,
+            sp_model_kwargs=self.sp_model_kwargs,
+            clean_up_tokenization_spaces=clean_up_tokenization_spaces,
+            **kwargs,
+        )
+
+    def __getstate__(self):
+        state = self.__dict__.copy()
+        state["sp_model"] = None
+        return state
+
+    def __setstate__(self, d):
+        self.__dict__ = d
+        self.sp_model = spm.SentencePieceProcessor(**self.sp_model_kwargs)
+        self.sp_model.Load(self.vocab_file)
+
+    @property
+    def vocab_size(self):
+        """Returns vocab size"""
+        return self.sp_model.get_piece_size()
+
+    def get_vocab(self):
+        """Returns vocab as a dict"""
+        vocab = {self.convert_ids_to_tokens(i): i for i in range(self.vocab_size)}
+        vocab.update(self.added_tokens_encoder)
+        return vocab
+
+    def _tokenize(self, text):
+        """Returns a tokenized string."""
+        return self.sp_model.encode(text, out_type=str)
+
+    def _convert_token_to_id(self, token):
+        """Converts a token (str) in an id using the vocab."""
+        return self.sp_model.piece_to_id(token)
+
+    def _convert_id_to_token(self, index):
+        """Converts an index (integer) in a token (str) using the vocab."""
+        token = self.sp_model.IdToPiece(index)
+        return token
+
+    def convert_tokens_to_string(self, tokens):
+        """Converts a sequence of tokens (string) in a single string."""
+        current_sub_tokens = []
+        out_string = ""
+        prev_is_special = False
+        for i, token in enumerate(tokens):
+            # make sure that special tokens are not decoded using sentencepiece model
+            if token in self.all_special_tokens:
+                if not prev_is_special and i != 0:
+                    out_string += " "
+                out_string += self.sp_model.decode(current_sub_tokens) + token
+                prev_is_special = True
+                current_sub_tokens = []
+            else:
+                current_sub_tokens.append(token)
+                prev_is_special = False
+        out_string += self.sp_model.decode(current_sub_tokens)
+        return out_string
+
+    def save_vocabulary(self, save_directory, filename_prefix: Optional[str] = None) -> Tuple[str]:
+        """
+        Save the vocabulary and special tokens file to a directory.
+
+        Args:
+            save_directory (`str`):
+                The directory in which to save the vocabulary.
+
+        Returns:
+            `Tuple(str)`: Paths to the files saved.
+        """
+        if not os.path.isdir(save_directory):
+            logger.error(f"Vocabulary path ({save_directory}) should be a directory")
+            return
+        out_vocab_file = os.path.join(
+            save_directory, (filename_prefix + "-" if filename_prefix else "") + VOCAB_FILES_NAMES["vocab_file"]
+        )
+
+        if os.path.abspath(self.vocab_file) != os.path.abspath(out_vocab_file) and os.path.isfile(self.vocab_file):
+            copyfile(self.vocab_file, out_vocab_file)
+        elif not os.path.isfile(self.vocab_file):
+            with open(out_vocab_file, "wb") as fi:
+                content_spiece_model = self.sp_model.serialized_model_proto()
+                fi.write(content_spiece_model)
+
+        return (out_vocab_file,)
+
+    def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=None):
+        bos_token_id = [self.bos_token_id] if self.add_bos_token else []
+        eos_token_id = [self.eos_token_id] if self.add_eos_token else []
+
+        output = bos_token_id + token_ids_0 + eos_token_id
+
+        if token_ids_1 is not None:
+            output = output + bos_token_id + token_ids_1 + eos_token_id
+
+        return output
+
+    def get_special_tokens_mask(
+        self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None, already_has_special_tokens: bool = False
+    ) -> List[int]:
+        """
+        Retrieve sequence ids from a token list that has no special tokens added. This method is called when adding
+        special tokens using the tokenizer `prepare_for_model` method.
+
+        Args:
+            token_ids_0 (`List[int]`):
+                List of IDs.
+            token_ids_1 (`List[int]`, *optional*):
+                Optional second list of IDs for sequence pairs.
+            already_has_special_tokens (`bool`, *optional*, defaults to `False`):
+                Whether or not the token list is already formatted with special tokens for the model.
+
+        Returns:
+            `List[int]`: A list of integers in the range [0, 1]: 1 for a special token, 0 for a sequence token.
+        """
+        if already_has_special_tokens:
+            return super().get_special_tokens_mask(
+                token_ids_0=token_ids_0, token_ids_1=token_ids_1, already_has_special_tokens=True
+            )
+
+        bos_token_id = [1] if self.add_bos_token else []
+        eos_token_id = [1] if self.add_eos_token else []
+
+        if token_ids_1 is None:
+            return bos_token_id + ([0] * len(token_ids_0)) + eos_token_id
+        return (
+            bos_token_id
+            + ([0] * len(token_ids_0))
+            + eos_token_id
+            + bos_token_id
+            + ([0] * len(token_ids_1))
+            + eos_token_id
+        )
+
+    def create_token_type_ids_from_sequences(
+        self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None
+    ) -> List[int]:
+        """
+        Creates a mask from the two sequences passed to be used in a sequence-pair classification task. An ALBERT
+        sequence pair mask has the following format:
+
+        ```
+        0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1
+        | first sequence    | second sequence |
+        ```
+
+        if token_ids_1 is None, only returns the first portion of the mask (0s).
+
+        Args:
+            token_ids_0 (`List[int]`):
+                List of ids.
+            token_ids_1 (`List[int]`, *optional*):
+                Optional second list of IDs for sequence pairs.
+
+        Returns:
+            `List[int]`: List of [token type IDs](../glossary#token-type-ids) according to the given sequence(s).
+        """
+        bos_token_id = [self.bos_token_id] if self.add_bos_token else []
+        eos_token_id = [self.eos_token_id] if self.add_eos_token else []
+
+        output = [0] * len(bos_token_id + token_ids_0 + eos_token_id)
+
+        if token_ids_1 is not None:
+            output += [1] * len(bos_token_id + token_ids_1 + eos_token_id)
+
+        return output

checkpoint-300/tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4be54af290d93c113bcbf421115ae9eed9d6340408f564898f1e966dc738ef01
+size 1136699

checkpoint-300/tokenizer_config.json

@@ -0,0 +1,47 @@
+{
+  "add_bos_token": false,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "auto_map": {
+    "AutoTokenizer": [
+      "tokenization_baichuan.BaiChuanTokenizer",
+      null
+    ]
+  },
+  "bos_token": "<s>",
+  "chat_template": "{% if messages[0]['role'] == 'system' %}{% set system_message = messages[0]['content'] %}{% endif %}{% if system_message is defined %}{{ system_message + '\\n' }}{% endif %}{% for message in messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ 'Human: ' + content + '\\nAssistant: ' }}{% elif message['role'] == 'assistant' %}{{ content + '</s>' + '\\n' }}{% endif %}{% endfor %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "</s>",
+  "padding_side": "right",
+  "sp_model_kwargs": {},
+  "split_special_tokens": false,
+  "tokenizer_class": "BaiChuanTokenizer",
+  "unk_token": "<unk>"
+}

checkpoint-300/trainer_state.json

@@ -0,0 +1,381 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.4,
+  "eval_steps": 500,
+  "global_step": 300,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.04,
+      "learning_rate": 0.00019991228300988585,
+      "loss": 2.1829,
+      "step": 5
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 0.00019964928592495045,
+      "loss": 1.8983,
+      "step": 10
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 0.0001992114701314478,
+      "loss": 1.7871,
+      "step": 15
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 0.0001985996037070505,
+      "loss": 1.8482,
+      "step": 20
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 0.00019781476007338058,
+      "loss": 1.731,
+      "step": 25
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 0.0001968583161128631,
+      "loss": 1.6797,
+      "step": 30
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 0.00019573194975320673,
+      "loss": 1.6694,
+      "step": 35
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 0.00019443763702374812,
+      "loss": 1.7493,
+      "step": 40
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 0.00019297764858882514,
+      "loss": 1.7823,
+      "step": 45
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 0.0001913545457642601,
+      "loss": 1.6518,
+      "step": 50
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 0.0001895711760239413,
+      "loss": 1.7334,
+      "step": 55
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 0.00018763066800438636,
+      "loss": 1.7565,
+      "step": 60
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 0.00018553642601605068,
+      "loss": 1.6493,
+      "step": 65
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 0.00018329212407100994,
+      "loss": 1.7431,
+      "step": 70
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 0.00018090169943749476,
+      "loss": 1.7561,
+      "step": 75
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 0.000178369345732584,
+      "loss": 1.6185,
+      "step": 80
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 0.00017569950556517566,
+      "loss": 1.6059,
+      "step": 85
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 0.00017289686274214118,
+      "loss": 1.772,
+      "step": 90
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 0.00016996633405133655,
+      "loss": 1.5847,
+      "step": 95
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 0.00016691306063588583,
+      "loss": 1.615,
+      "step": 100
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 0.000163742398974869,
+      "loss": 1.5706,
+      "step": 105
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 0.0001604599114862375,
+      "loss": 1.6997,
+      "step": 110
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 0.0001570713567684432,
+      "loss": 1.5529,
+      "step": 115
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 0.00015358267949789966,
+      "loss": 1.6631,
+      "step": 120
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 0.00015000000000000001,
+      "loss": 1.7483,
+      "step": 125
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 0.00014632960351198618,
+      "loss": 1.708,
+      "step": 130
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 0.00014257792915650728,
+      "loss": 1.6979,
+      "step": 135
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 0.0001387515586452103,
+      "loss": 1.53,
+      "step": 140
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 0.00013485720473218154,
+      "loss": 1.6821,
+      "step": 145
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 0.00013090169943749476,
+      "loss": 1.7208,
+      "step": 150
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 0.00012689198206152657,
+      "loss": 1.6841,
+      "step": 155
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 0.00012283508701106557,
+      "loss": 1.544,
+      "step": 160
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 0.00011873813145857249,
+      "loss": 1.5851,
+      "step": 165
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 0.00011460830285624118,
+      "loss": 1.56,
+      "step": 170
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 0.00011045284632676536,
+      "loss": 1.5691,
+      "step": 175
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 0.00010627905195293135,
+      "loss": 1.5201,
+      "step": 180
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 0.0001020942419883357,
+      "loss": 1.5098,
+      "step": 185
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 9.790575801166432e-05,
+      "loss": 1.5805,
+      "step": 190
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 9.372094804706867e-05,
+      "loss": 1.6742,
+      "step": 195
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 8.954715367323468e-05,
+      "loss": 1.5656,
+      "step": 200
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 8.539169714375885e-05,
+      "loss": 1.6301,
+      "step": 205
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 8.126186854142752e-05,
+      "loss": 1.6027,
+      "step": 210
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 7.716491298893442e-05,
+      "loss": 1.6494,
+      "step": 215
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 7.310801793847344e-05,
+      "loss": 1.5962,
+      "step": 220
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 6.909830056250527e-05,
+      "loss": 1.5375,
+      "step": 225
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 6.51427952678185e-05,
+      "loss": 1.596,
+      "step": 230
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 6.12484413547897e-05,
+      "loss": 1.6401,
+      "step": 235
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 5.7422070843492734e-05,
+      "loss": 1.5735,
+      "step": 240
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 5.3670396488013854e-05,
+      "loss": 1.6057,
+      "step": 245
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 5.000000000000002e-05,
+      "loss": 1.5428,
+      "step": 250
+    },
+    {
+      "epoch": 2.04,
+      "learning_rate": 4.6417320502100316e-05,
+      "loss": 1.6843,
+      "step": 255
+    },
+    {
+      "epoch": 2.08,
+      "learning_rate": 4.2928643231556844e-05,
+      "loss": 1.6004,
+      "step": 260
+    },
+    {
+      "epoch": 2.12,
+      "learning_rate": 3.954008851376252e-05,
+      "loss": 1.5231,
+      "step": 265
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 3.6257601025131026e-05,
+      "loss": 1.6147,
+      "step": 270
+    },
+    {
+      "epoch": 2.2,
+      "learning_rate": 3.308693936411421e-05,
+      "loss": 1.5095,
+      "step": 275
+    },
+    {
+      "epoch": 2.24,
+      "learning_rate": 3.0033665948663448e-05,
+      "loss": 1.6355,
+      "step": 280
+    },
+    {
+      "epoch": 2.28,
+      "learning_rate": 2.7103137257858868e-05,
+      "loss": 1.4353,
+      "step": 285
+    },
+    {
+      "epoch": 2.32,
+      "learning_rate": 2.4300494434824373e-05,
+      "loss": 1.5886,
+      "step": 290
+    },
+    {
+      "epoch": 2.36,
+      "learning_rate": 2.163065426741603e-05,
+      "loss": 1.5915,
+      "step": 295
+    },
+    {
+      "epoch": 2.4,
+      "learning_rate": 1.9098300562505266e-05,
+      "loss": 1.5584,
+      "step": 300
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 375,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 100,
+  "total_flos": 3.46235716435968e+16,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-300/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:84990e28d454e290f6393201e221ba051e01af18e581a4f5994ac8396ad7c48b
+size 4920

special_tokens_map.json

@@ -0,0 +1,24 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "</s>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenization_baichuan.py

@@ -0,0 +1,251 @@
+# coding=utf-8
+# Copyright 2022 EleutherAI and the HuggingFace Inc. team. All rights reserved.
+#
+# This code is based on EleutherAI's GPT-NeoX library and the GPT-NeoX
+# and OPT implementations in this library. It has been modified from its
+# original forms to accommodate minor architectural differences compared
+# to GPT-NeoX and OPT used by the Meta AI team that trained the model.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import os
+from shutil import copyfile
+from typing import Any, Dict, List, Optional, Tuple
+
+import sentencepiece as spm
+
+from transformers.tokenization_utils import AddedToken, PreTrainedTokenizer
+from transformers.utils import logging
+
+
+logger = logging.get_logger(__name__)
+
+VOCAB_FILES_NAMES = {"vocab_file": "tokenizer.model"}
+
+PRETRAINED_VOCAB_FILES_MAP = {
+    "vocab_file": {},
+    "tokenizer_file": {},
+}
+PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES = {}
+
+
+class BaiChuanTokenizer(PreTrainedTokenizer):
+    """
+    Construct a BaiChuan tokenizer. Based on byte-level Byte-Pair-Encoding.
+
+    Args:
+        vocab_file (`str`):
+            Path to the vocabulary file.
+    """
+
+    vocab_files_names = VOCAB_FILES_NAMES
+    pretrained_vocab_files_map = PRETRAINED_VOCAB_FILES_MAP
+    max_model_input_sizes = PRETRAINED_POSITIONAL_EMBEDDINGS_SIZES
+    model_input_names = ["input_ids", "attention_mask"]
+
+    def __init__(
+        self,
+        vocab_file,
+        unk_token="<unk>",
+        bos_token="<s>",
+        eos_token="</s>",
+        pad_token=None,
+        sp_model_kwargs: Optional[Dict[str, Any]] = None,
+        add_bos_token=True,
+        add_eos_token=False,
+        clean_up_tokenization_spaces=False,
+        **kwargs,
+    ):
+        self.sp_model_kwargs = {} if sp_model_kwargs is None else sp_model_kwargs
+        bos_token = AddedToken(bos_token, lstrip=False, rstrip=False) if isinstance(bos_token, str) else bos_token
+        eos_token = AddedToken(eos_token, lstrip=False, rstrip=False) if isinstance(eos_token, str) else eos_token
+        unk_token = AddedToken(unk_token, lstrip=False, rstrip=False) if isinstance(unk_token, str) else unk_token
+        pad_token = AddedToken(pad_token, lstrip=False, rstrip=False) if isinstance(pad_token, str) else pad_token
+        self.vocab_file = vocab_file
+        self.add_bos_token = add_bos_token
+        self.add_eos_token = add_eos_token
+        self.sp_model = spm.SentencePieceProcessor(**self.sp_model_kwargs)
+        self.sp_model.Load(vocab_file)
+        
+        super().__init__(
+            bos_token=bos_token,
+            eos_token=eos_token,
+            unk_token=unk_token,
+            pad_token=pad_token,
+            add_bos_token=add_bos_token,
+            add_eos_token=add_eos_token,
+            sp_model_kwargs=self.sp_model_kwargs,
+            clean_up_tokenization_spaces=clean_up_tokenization_spaces,
+            **kwargs,
+        )
+
+    def __getstate__(self):
+        state = self.__dict__.copy()
+        state["sp_model"] = None
+        return state
+
+    def __setstate__(self, d):
+        self.__dict__ = d
+        self.sp_model = spm.SentencePieceProcessor(**self.sp_model_kwargs)
+        self.sp_model.Load(self.vocab_file)
+
+    @property
+    def vocab_size(self):
+        """Returns vocab size"""
+        return self.sp_model.get_piece_size()
+
+    def get_vocab(self):
+        """Returns vocab as a dict"""
+        vocab = {self.convert_ids_to_tokens(i): i for i in range(self.vocab_size)}
+        vocab.update(self.added_tokens_encoder)
+        return vocab
+
+    def _tokenize(self, text):
+        """Returns a tokenized string."""
+        return self.sp_model.encode(text, out_type=str)
+
+    def _convert_token_to_id(self, token):
+        """Converts a token (str) in an id using the vocab."""
+        return self.sp_model.piece_to_id(token)
+
+    def _convert_id_to_token(self, index):
+        """Converts an index (integer) in a token (str) using the vocab."""
+        token = self.sp_model.IdToPiece(index)
+        return token
+
+    def convert_tokens_to_string(self, tokens):
+        """Converts a sequence of tokens (string) in a single string."""
+        current_sub_tokens = []
+        out_string = ""
+        prev_is_special = False
+        for i, token in enumerate(tokens):
+            # make sure that special tokens are not decoded using sentencepiece model
+            if token in self.all_special_tokens:
+                if not prev_is_special and i != 0:
+                    out_string += " "
+                out_string += self.sp_model.decode(current_sub_tokens) + token
+                prev_is_special = True
+                current_sub_tokens = []
+            else:
+                current_sub_tokens.append(token)
+                prev_is_special = False
+        out_string += self.sp_model.decode(current_sub_tokens)
+        return out_string
+
+    def save_vocabulary(self, save_directory, filename_prefix: Optional[str] = None) -> Tuple[str]:
+        """
+        Save the vocabulary and special tokens file to a directory.
+
+        Args:
+            save_directory (`str`):
+                The directory in which to save the vocabulary.
+
+        Returns:
+            `Tuple(str)`: Paths to the files saved.
+        """
+        if not os.path.isdir(save_directory):
+            logger.error(f"Vocabulary path ({save_directory}) should be a directory")
+            return
+        out_vocab_file = os.path.join(
+            save_directory, (filename_prefix + "-" if filename_prefix else "") + VOCAB_FILES_NAMES["vocab_file"]
+        )
+
+        if os.path.abspath(self.vocab_file) != os.path.abspath(out_vocab_file) and os.path.isfile(self.vocab_file):
+            copyfile(self.vocab_file, out_vocab_file)
+        elif not os.path.isfile(self.vocab_file):
+            with open(out_vocab_file, "wb") as fi:
+                content_spiece_model = self.sp_model.serialized_model_proto()
+                fi.write(content_spiece_model)
+
+        return (out_vocab_file,)
+
+    def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=None):
+        bos_token_id = [self.bos_token_id] if self.add_bos_token else []
+        eos_token_id = [self.eos_token_id] if self.add_eos_token else []
+
+        output = bos_token_id + token_ids_0 + eos_token_id
+
+        if token_ids_1 is not None:
+            output = output + bos_token_id + token_ids_1 + eos_token_id
+
+        return output
+
+    def get_special_tokens_mask(
+        self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None, already_has_special_tokens: bool = False
+    ) -> List[int]:
+        """
+        Retrieve sequence ids from a token list that has no special tokens added. This method is called when adding
+        special tokens using the tokenizer `prepare_for_model` method.
+
+        Args:
+            token_ids_0 (`List[int]`):
+                List of IDs.
+            token_ids_1 (`List[int]`, *optional*):
+                Optional second list of IDs for sequence pairs.
+            already_has_special_tokens (`bool`, *optional*, defaults to `False`):
+                Whether or not the token list is already formatted with special tokens for the model.
+
+        Returns:
+            `List[int]`: A list of integers in the range [0, 1]: 1 for a special token, 0 for a sequence token.
+        """
+        if already_has_special_tokens:
+            return super().get_special_tokens_mask(
+                token_ids_0=token_ids_0, token_ids_1=token_ids_1, already_has_special_tokens=True
+            )
+
+        bos_token_id = [1] if self.add_bos_token else []
+        eos_token_id = [1] if self.add_eos_token else []
+
+        if token_ids_1 is None:
+            return bos_token_id + ([0] * len(token_ids_0)) + eos_token_id
+        return (
+            bos_token_id
+            + ([0] * len(token_ids_0))
+            + eos_token_id
+            + bos_token_id
+            + ([0] * len(token_ids_1))
+            + eos_token_id
+        )
+
+    def create_token_type_ids_from_sequences(
+        self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None
+    ) -> List[int]:
+        """
+        Creates a mask from the two sequences passed to be used in a sequence-pair classification task. An ALBERT
+        sequence pair mask has the following format:
+
+        ```
+        0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1
+        | first sequence    | second sequence |
+        ```
+
+        if token_ids_1 is None, only returns the first portion of the mask (0s).
+
+        Args:
+            token_ids_0 (`List[int]`):
+                List of ids.
+            token_ids_1 (`List[int]`, *optional*):
+                Optional second list of IDs for sequence pairs.
+
+        Returns:
+            `List[int]`: List of [token type IDs](../glossary#token-type-ids) according to the given sequence(s).
+        """
+        bos_token_id = [self.bos_token_id] if self.add_bos_token else []
+        eos_token_id = [self.eos_token_id] if self.add_eos_token else []
+
+        output = [0] * len(bos_token_id + token_ids_0 + eos_token_id)
+
+        if token_ids_1 is not None:
+            output += [1] * len(bos_token_id + token_ids_1 + eos_token_id)
+
+        return output

tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4be54af290d93c113bcbf421115ae9eed9d6340408f564898f1e966dc738ef01
+size 1136699

tokenizer_config.json

@@ -0,0 +1,47 @@
+{
+  "add_bos_token": false,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "auto_map": {
+    "AutoTokenizer": [
+      "tokenization_baichuan.BaiChuanTokenizer",
+      null
+    ]
+  },
+  "bos_token": "<s>",
+  "chat_template": "{% if messages[0]['role'] == 'system' %}{% set system_message = messages[0]['content'] %}{% endif %}{% if system_message is defined %}{{ system_message + '\\n' }}{% endif %}{% for message in messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ 'Human: ' + content + '\\nAssistant: ' }}{% elif message['role'] == 'assistant' %}{{ content + '</s>' + '\\n' }}{% endif %}{% endfor %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "</s>",
+  "padding_side": "right",
+  "sp_model_kwargs": {},
+  "split_special_tokens": false,
+  "tokenizer_class": "BaiChuanTokenizer",
+  "unk_token": "<unk>"
+}

train_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 3.0,
+    "train_loss": 1.628477378845215,
+    "train_runtime": 1045.5144,
+    "train_samples_per_second": 5.739,
+    "train_steps_per_second": 0.359
+}

trainer_log.jsonl

@@ -0,0 +1,76 @@
+{"current_steps": 5, "total_steps": 375, "loss": 2.1829, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 0.00019991228300988585, "epoch": 0.04, "percentage": 1.33, "elapsed_time": "0:00:12", "remaining_time": "0:15:11"}
+{"current_steps": 10, "total_steps": 375, "loss": 1.8983, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 0.00019964928592495045, "epoch": 0.08, "percentage": 2.67, "elapsed_time": "0:00:25", "remaining_time": "0:15:39"}
+{"current_steps": 15, "total_steps": 375, "loss": 1.7871, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 0.0001992114701314478, "epoch": 0.12, "percentage": 4.0, "elapsed_time": "0:00:38", "remaining_time": "0:15:27"}
+{"current_steps": 20, "total_steps": 375, "loss": 1.8482, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 0.0001985996037070505, "epoch": 0.16, "percentage": 5.33, "elapsed_time": "0:00:52", "remaining_time": "0:15:39"}
+{"current_steps": 25, "total_steps": 375, "loss": 1.731, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 0.00019781476007338058, "epoch": 0.2, "percentage": 6.67, "elapsed_time": "0:01:07", "remaining_time": "0:15:48"}
+{"current_steps": 30, "total_steps": 375, "loss": 1.6797, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 0.0001968583161128631, "epoch": 0.24, "percentage": 8.0, "elapsed_time": "0:01:22", "remaining_time": "0:15:52"}
+{"current_steps": 35, "total_steps": 375, "loss": 1.6694, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 0.00019573194975320673, "epoch": 0.28, "percentage": 9.33, "elapsed_time": "0:01:36", "remaining_time": "0:15:32"}
+{"current_steps": 40, "total_steps": 375, "loss": 1.7493, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 0.00019443763702374812, "epoch": 0.32, "percentage": 10.67, "elapsed_time": "0:01:50", "remaining_time": "0:15:24"}
+{"current_steps": 45, "total_steps": 375, "loss": 1.7823, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 0.00019297764858882514, "epoch": 0.36, "percentage": 12.0, "elapsed_time": "0:02:03", "remaining_time": "0:15:08"}
+{"current_steps": 50, "total_steps": 375, "loss": 1.6518, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 0.0001913545457642601, "epoch": 0.4, "percentage": 13.33, "elapsed_time": "0:02:16", "remaining_time": "0:14:45"}
+{"current_steps": 55, "total_steps": 375, "loss": 1.7334, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 0.0001895711760239413, "epoch": 0.44, "percentage": 14.67, "elapsed_time": "0:02:30", "remaining_time": "0:14:37"}
+{"current_steps": 60, "total_steps": 375, "loss": 1.7565, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 0.00018763066800438636, "epoch": 0.48, "percentage": 16.0, "elapsed_time": "0:02:43", "remaining_time": "0:14:18"}
+{"current_steps": 65, "total_steps": 375, "loss": 1.6493, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 0.00018553642601605068, "epoch": 0.52, "percentage": 17.33, "elapsed_time": "0:02:56", "remaining_time": "0:14:02"}
+{"current_steps": 70, "total_steps": 375, "loss": 1.7431, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 0.00018329212407100994, "epoch": 0.56, "percentage": 18.67, "elapsed_time": "0:03:10", "remaining_time": "0:13:47"}
+{"current_steps": 75, "total_steps": 375, "loss": 1.7561, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 0.00018090169943749476, "epoch": 0.6, "percentage": 20.0, "elapsed_time": "0:03:23", "remaining_time": "0:13:32"}
+{"current_steps": 80, "total_steps": 375, "loss": 1.6185, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 0.000178369345732584, "epoch": 0.64, "percentage": 21.33, "elapsed_time": "0:03:39", "remaining_time": "0:13:30"}
+{"current_steps": 85, "total_steps": 375, "loss": 1.6059, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 0.00017569950556517566, "epoch": 0.68, "percentage": 22.67, "elapsed_time": "0:03:54", "remaining_time": "0:13:18"}
+{"current_steps": 90, "total_steps": 375, "loss": 1.772, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 0.00017289686274214118, "epoch": 0.72, "percentage": 24.0, "elapsed_time": "0:04:09", "remaining_time": "0:13:10"}
+{"current_steps": 95, "total_steps": 375, "loss": 1.5847, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 0.00016996633405133655, "epoch": 0.76, "percentage": 25.33, "elapsed_time": "0:04:23", "remaining_time": "0:12:57"}
+{"current_steps": 100, "total_steps": 375, "loss": 1.615, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 0.00016691306063588583, "epoch": 0.8, "percentage": 26.67, "elapsed_time": "0:04:37", "remaining_time": "0:12:42"}
+{"current_steps": 105, "total_steps": 375, "loss": 1.5706, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 0.000163742398974869, "epoch": 0.84, "percentage": 28.0, "elapsed_time": "0:04:51", "remaining_time": "0:12:29"}
+{"current_steps": 110, "total_steps": 375, "loss": 1.6997, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 0.0001604599114862375, "epoch": 0.88, "percentage": 29.33, "elapsed_time": "0:05:04", "remaining_time": "0:12:14"}
+{"current_steps": 115, "total_steps": 375, "loss": 1.5529, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 0.0001570713567684432, "epoch": 0.92, "percentage": 30.67, "elapsed_time": "0:05:19", "remaining_time": "0:12:02"}
+{"current_steps": 120, "total_steps": 375, "loss": 1.6631, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 0.00015358267949789966, "epoch": 0.96, "percentage": 32.0, "elapsed_time": "0:05:33", "remaining_time": "0:11:48"}
+{"current_steps": 125, "total_steps": 375, "loss": 1.7483, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 0.00015000000000000001, "epoch": 1.0, "percentage": 33.33, "elapsed_time": "0:05:47", "remaining_time": "0:11:34"}
+{"current_steps": 130, "total_steps": 375, "loss": 1.708, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 0.00014632960351198618, "epoch": 1.04, "percentage": 34.67, "elapsed_time": "0:06:02", "remaining_time": "0:11:22"}
+{"current_steps": 135, "total_steps": 375, "loss": 1.6979, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 0.00014257792915650728, "epoch": 1.08, "percentage": 36.0, "elapsed_time": "0:06:16", "remaining_time": "0:11:09"}
+{"current_steps": 140, "total_steps": 375, "loss": 1.53, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 0.0001387515586452103, "epoch": 1.12, "percentage": 37.33, "elapsed_time": "0:06:31", "remaining_time": "0:10:57"}
+{"current_steps": 145, "total_steps": 375, "loss": 1.6821, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 0.00013485720473218154, "epoch": 1.16, "percentage": 38.67, "elapsed_time": "0:06:44", "remaining_time": "0:10:41"}
+{"current_steps": 150, "total_steps": 375, "loss": 1.7208, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 0.00013090169943749476, "epoch": 1.2, "percentage": 40.0, "elapsed_time": "0:06:58", "remaining_time": "0:10:27"}
+{"current_steps": 155, "total_steps": 375, "loss": 1.6841, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 0.00012689198206152657, "epoch": 1.24, "percentage": 41.33, "elapsed_time": "0:07:13", "remaining_time": "0:10:14"}
+{"current_steps": 160, "total_steps": 375, "loss": 1.544, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 0.00012283508701106557, "epoch": 1.28, "percentage": 42.67, "elapsed_time": "0:07:27", "remaining_time": "0:10:00"}
+{"current_steps": 165, "total_steps": 375, "loss": 1.5851, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 0.00011873813145857249, "epoch": 1.32, "percentage": 44.0, "elapsed_time": "0:07:40", "remaining_time": "0:09:46"}
+{"current_steps": 170, "total_steps": 375, "loss": 1.56, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 0.00011460830285624118, "epoch": 1.36, "percentage": 45.33, "elapsed_time": "0:07:54", "remaining_time": "0:09:32"}
+{"current_steps": 175, "total_steps": 375, "loss": 1.5691, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 0.00011045284632676536, "epoch": 1.4, "percentage": 46.67, "elapsed_time": "0:08:09", "remaining_time": "0:09:19"}
+{"current_steps": 180, "total_steps": 375, "loss": 1.5201, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 0.00010627905195293135, "epoch": 1.44, "percentage": 48.0, "elapsed_time": "0:08:22", "remaining_time": "0:09:04"}
+{"current_steps": 185, "total_steps": 375, "loss": 1.5098, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 0.0001020942419883357, "epoch": 1.48, "percentage": 49.33, "elapsed_time": "0:08:37", "remaining_time": "0:08:51"}
+{"current_steps": 190, "total_steps": 375, "loss": 1.5805, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 9.790575801166432e-05, "epoch": 1.52, "percentage": 50.67, "elapsed_time": "0:08:52", "remaining_time": "0:08:38"}
+{"current_steps": 195, "total_steps": 375, "loss": 1.6742, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 9.372094804706867e-05, "epoch": 1.56, "percentage": 52.0, "elapsed_time": "0:09:05", "remaining_time": "0:08:23"}
+{"current_steps": 200, "total_steps": 375, "loss": 1.5656, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 8.954715367323468e-05, "epoch": 1.6, "percentage": 53.33, "elapsed_time": "0:09:18", "remaining_time": "0:08:08"}
+{"current_steps": 205, "total_steps": 375, "loss": 1.6301, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 8.539169714375885e-05, "epoch": 1.64, "percentage": 54.67, "elapsed_time": "0:09:33", "remaining_time": "0:07:55"}
+{"current_steps": 210, "total_steps": 375, "loss": 1.6027, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 8.126186854142752e-05, "epoch": 1.68, "percentage": 56.0, "elapsed_time": "0:09:47", "remaining_time": "0:07:41"}
+{"current_steps": 215, "total_steps": 375, "loss": 1.6494, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 7.716491298893442e-05, "epoch": 1.72, "percentage": 57.33, "elapsed_time": "0:10:01", "remaining_time": "0:07:27"}
+{"current_steps": 220, "total_steps": 375, "loss": 1.5962, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 7.310801793847344e-05, "epoch": 1.76, "percentage": 58.67, "elapsed_time": "0:10:16", "remaining_time": "0:07:14"}
+{"current_steps": 225, "total_steps": 375, "loss": 1.5375, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 6.909830056250527e-05, "epoch": 1.8, "percentage": 60.0, "elapsed_time": "0:10:30", "remaining_time": "0:07:00"}
+{"current_steps": 230, "total_steps": 375, "loss": 1.596, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 6.51427952678185e-05, "epoch": 1.84, "percentage": 61.33, "elapsed_time": "0:10:43", "remaining_time": "0:06:45"}
+{"current_steps": 235, "total_steps": 375, "loss": 1.6401, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 6.12484413547897e-05, "epoch": 1.88, "percentage": 62.67, "elapsed_time": "0:10:57", "remaining_time": "0:06:31"}
+{"current_steps": 240, "total_steps": 375, "loss": 1.5735, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 5.7422070843492734e-05, "epoch": 1.92, "percentage": 64.0, "elapsed_time": "0:11:11", "remaining_time": "0:06:17"}
+{"current_steps": 245, "total_steps": 375, "loss": 1.6057, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 5.3670396488013854e-05, "epoch": 1.96, "percentage": 65.33, "elapsed_time": "0:11:23", "remaining_time": "0:06:02"}
+{"current_steps": 250, "total_steps": 375, "loss": 1.5428, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 5.000000000000002e-05, "epoch": 2.0, "percentage": 66.67, "elapsed_time": "0:11:37", "remaining_time": "0:05:48"}
+{"current_steps": 255, "total_steps": 375, "loss": 1.6843, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.6417320502100316e-05, "epoch": 2.04, "percentage": 68.0, "elapsed_time": "0:11:51", "remaining_time": "0:05:34"}
+{"current_steps": 260, "total_steps": 375, "loss": 1.6004, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.2928643231556844e-05, "epoch": 2.08, "percentage": 69.33, "elapsed_time": "0:12:05", "remaining_time": "0:05:20"}
+{"current_steps": 265, "total_steps": 375, "loss": 1.5231, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.954008851376252e-05, "epoch": 2.12, "percentage": 70.67, "elapsed_time": "0:12:18", "remaining_time": "0:05:06"}
+{"current_steps": 270, "total_steps": 375, "loss": 1.6147, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.6257601025131026e-05, "epoch": 2.16, "percentage": 72.0, "elapsed_time": "0:12:32", "remaining_time": "0:04:52"}
+{"current_steps": 275, "total_steps": 375, "loss": 1.5095, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.308693936411421e-05, "epoch": 2.2, "percentage": 73.33, "elapsed_time": "0:12:47", "remaining_time": "0:04:38"}
+{"current_steps": 280, "total_steps": 375, "loss": 1.6355, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.0033665948663448e-05, "epoch": 2.24, "percentage": 74.67, "elapsed_time": "0:13:01", "remaining_time": "0:04:25"}
+{"current_steps": 285, "total_steps": 375, "loss": 1.4353, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.7103137257858868e-05, "epoch": 2.28, "percentage": 76.0, "elapsed_time": "0:13:12", "remaining_time": "0:04:10"}
+{"current_steps": 290, "total_steps": 375, "loss": 1.5886, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.4300494434824373e-05, "epoch": 2.32, "percentage": 77.33, "elapsed_time": "0:13:27", "remaining_time": "0:03:56"}
+{"current_steps": 295, "total_steps": 375, "loss": 1.5915, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.163065426741603e-05, "epoch": 2.36, "percentage": 78.67, "elapsed_time": "0:13:42", "remaining_time": "0:03:42"}
+{"current_steps": 300, "total_steps": 375, "loss": 1.5584, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.9098300562505266e-05, "epoch": 2.4, "percentage": 80.0, "elapsed_time": "0:13:56", "remaining_time": "0:03:29"}
+{"current_steps": 305, "total_steps": 375, "loss": 1.4628, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.6707875928990058e-05, "epoch": 2.44, "percentage": 81.33, "elapsed_time": "0:14:11", "remaining_time": "0:03:15"}
+{"current_steps": 310, "total_steps": 375, "loss": 1.5502, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.4463573983949341e-05, "epoch": 2.48, "percentage": 82.67, "elapsed_time": "0:14:27", "remaining_time": "0:03:01"}
+{"current_steps": 315, "total_steps": 375, "loss": 1.4882, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.2369331995613665e-05, "epoch": 2.52, "percentage": 84.0, "elapsed_time": "0:14:41", "remaining_time": "0:02:47"}
+{"current_steps": 320, "total_steps": 375, "loss": 1.6666, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.042882397605871e-05, "epoch": 2.56, "percentage": 85.33, "elapsed_time": "0:14:55", "remaining_time": "0:02:33"}
+{"current_steps": 325, "total_steps": 375, "loss": 1.4874, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 8.645454235739903e-06, "epoch": 2.6, "percentage": 86.67, "elapsed_time": "0:15:08", "remaining_time": "0:02:19"}
+{"current_steps": 330, "total_steps": 375, "loss": 1.6158, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 7.022351411174866e-06, "epoch": 2.64, "percentage": 88.0, "elapsed_time": "0:15:20", "remaining_time": "0:02:05"}
+{"current_steps": 335, "total_steps": 375, "loss": 1.4376, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 5.562362976251901e-06, "epoch": 2.68, "percentage": 89.33, "elapsed_time": "0:15:33", "remaining_time": "0:01:51"}
+{"current_steps": 340, "total_steps": 375, "loss": 1.6202, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.268050246793276e-06, "epoch": 2.72, "percentage": 90.67, "elapsed_time": "0:15:48", "remaining_time": "0:01:37"}
+{"current_steps": 345, "total_steps": 375, "loss": 1.5493, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.1416838871368924e-06, "epoch": 2.76, "percentage": 92.0, "elapsed_time": "0:16:02", "remaining_time": "0:01:23"}
+{"current_steps": 350, "total_steps": 375, "loss": 1.6157, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.1852399266194314e-06, "epoch": 2.8, "percentage": 93.33, "elapsed_time": "0:16:17", "remaining_time": "0:01:09"}
+{"current_steps": 355, "total_steps": 375, "loss": 1.5631, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.400396292949513e-06, "epoch": 2.84, "percentage": 94.67, "elapsed_time": "0:16:32", "remaining_time": "0:00:55"}
+{"current_steps": 360, "total_steps": 375, "loss": 1.5326, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 7.885298685522235e-07, "epoch": 2.88, "percentage": 96.0, "elapsed_time": "0:16:43", "remaining_time": "0:00:41"}
+{"current_steps": 365, "total_steps": 375, "loss": 1.5978, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.50714075049563e-07, "epoch": 2.92, "percentage": 97.33, "elapsed_time": "0:16:57", "remaining_time": "0:00:27"}
+{"current_steps": 370, "total_steps": 375, "loss": 1.4251, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 8.771699011416168e-08, "epoch": 2.96, "percentage": 98.67, "elapsed_time": "0:17:12", "remaining_time": "0:00:13"}
+{"current_steps": 375, "total_steps": 375, "loss": 1.6276, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 0.0, "epoch": 3.0, "percentage": 100.0, "elapsed_time": "0:17:25", "remaining_time": "0:00:00"}
+{"current_steps": 375, "total_steps": 375, "loss": null, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 3.0, "percentage": 100.0, "elapsed_time": "0:17:25", "remaining_time": "0:00:00"}

trainer_state.json

@@ -0,0 +1,480 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 3.0,
+  "eval_steps": 500,
+  "global_step": 375,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.04,
+      "learning_rate": 0.00019991228300988585,
+      "loss": 2.1829,
+      "step": 5
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 0.00019964928592495045,
+      "loss": 1.8983,
+      "step": 10
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 0.0001992114701314478,
+      "loss": 1.7871,
+      "step": 15
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 0.0001985996037070505,
+      "loss": 1.8482,
+      "step": 20
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 0.00019781476007338058,
+      "loss": 1.731,
+      "step": 25
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 0.0001968583161128631,
+      "loss": 1.6797,
+      "step": 30
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 0.00019573194975320673,
+      "loss": 1.6694,
+      "step": 35
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 0.00019443763702374812,
+      "loss": 1.7493,
+      "step": 40
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 0.00019297764858882514,
+      "loss": 1.7823,
+      "step": 45
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 0.0001913545457642601,
+      "loss": 1.6518,
+      "step": 50
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 0.0001895711760239413,
+      "loss": 1.7334,
+      "step": 55
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 0.00018763066800438636,
+      "loss": 1.7565,
+      "step": 60
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 0.00018553642601605068,
+      "loss": 1.6493,
+      "step": 65
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 0.00018329212407100994,
+      "loss": 1.7431,
+      "step": 70
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 0.00018090169943749476,
+      "loss": 1.7561,
+      "step": 75
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 0.000178369345732584,
+      "loss": 1.6185,
+      "step": 80
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 0.00017569950556517566,
+      "loss": 1.6059,
+      "step": 85
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 0.00017289686274214118,
+      "loss": 1.772,
+      "step": 90
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 0.00016996633405133655,
+      "loss": 1.5847,
+      "step": 95
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 0.00016691306063588583,
+      "loss": 1.615,
+      "step": 100
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 0.000163742398974869,
+      "loss": 1.5706,
+      "step": 105
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 0.0001604599114862375,
+      "loss": 1.6997,
+      "step": 110
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 0.0001570713567684432,
+      "loss": 1.5529,
+      "step": 115
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 0.00015358267949789966,
+      "loss": 1.6631,
+      "step": 120
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 0.00015000000000000001,
+      "loss": 1.7483,
+      "step": 125
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 0.00014632960351198618,
+      "loss": 1.708,
+      "step": 130
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 0.00014257792915650728,
+      "loss": 1.6979,
+      "step": 135
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 0.0001387515586452103,
+      "loss": 1.53,
+      "step": 140
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 0.00013485720473218154,
+      "loss": 1.6821,
+      "step": 145
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 0.00013090169943749476,
+      "loss": 1.7208,
+      "step": 150
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 0.00012689198206152657,
+      "loss": 1.6841,
+      "step": 155
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 0.00012283508701106557,
+      "loss": 1.544,
+      "step": 160
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 0.00011873813145857249,
+      "loss": 1.5851,
+      "step": 165
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 0.00011460830285624118,
+      "loss": 1.56,
+      "step": 170
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 0.00011045284632676536,
+      "loss": 1.5691,
+      "step": 175
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 0.00010627905195293135,
+      "loss": 1.5201,
+      "step": 180
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 0.0001020942419883357,
+      "loss": 1.5098,
+      "step": 185
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 9.790575801166432e-05,
+      "loss": 1.5805,
+      "step": 190
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 9.372094804706867e-05,
+      "loss": 1.6742,
+      "step": 195
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 8.954715367323468e-05,
+      "loss": 1.5656,
+      "step": 200
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 8.539169714375885e-05,
+      "loss": 1.6301,
+      "step": 205
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 8.126186854142752e-05,
+      "loss": 1.6027,
+      "step": 210
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 7.716491298893442e-05,
+      "loss": 1.6494,
+      "step": 215
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 7.310801793847344e-05,
+      "loss": 1.5962,
+      "step": 220
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 6.909830056250527e-05,
+      "loss": 1.5375,
+      "step": 225
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 6.51427952678185e-05,
+      "loss": 1.596,
+      "step": 230
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 6.12484413547897e-05,
+      "loss": 1.6401,
+      "step": 235
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 5.7422070843492734e-05,
+      "loss": 1.5735,
+      "step": 240
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 5.3670396488013854e-05,
+      "loss": 1.6057,
+      "step": 245
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 5.000000000000002e-05,
+      "loss": 1.5428,
+      "step": 250
+    },
+    {
+      "epoch": 2.04,
+      "learning_rate": 4.6417320502100316e-05,
+      "loss": 1.6843,
+      "step": 255
+    },
+    {
+      "epoch": 2.08,
+      "learning_rate": 4.2928643231556844e-05,
+      "loss": 1.6004,
+      "step": 260
+    },
+    {
+      "epoch": 2.12,
+      "learning_rate": 3.954008851376252e-05,
+      "loss": 1.5231,
+      "step": 265
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 3.6257601025131026e-05,
+      "loss": 1.6147,
+      "step": 270
+    },
+    {
+      "epoch": 2.2,
+      "learning_rate": 3.308693936411421e-05,
+      "loss": 1.5095,
+      "step": 275
+    },
+    {
+      "epoch": 2.24,
+      "learning_rate": 3.0033665948663448e-05,
+      "loss": 1.6355,
+      "step": 280
+    },
+    {
+      "epoch": 2.28,
+      "learning_rate": 2.7103137257858868e-05,
+      "loss": 1.4353,
+      "step": 285
+    },
+    {
+      "epoch": 2.32,
+      "learning_rate": 2.4300494434824373e-05,
+      "loss": 1.5886,
+      "step": 290
+    },
+    {
+      "epoch": 2.36,
+      "learning_rate": 2.163065426741603e-05,
+      "loss": 1.5915,
+      "step": 295
+    },
+    {
+      "epoch": 2.4,
+      "learning_rate": 1.9098300562505266e-05,
+      "loss": 1.5584,
+      "step": 300
+    },
+    {
+      "epoch": 2.44,
+      "learning_rate": 1.6707875928990058e-05,
+      "loss": 1.4628,
+      "step": 305
+    },
+    {
+      "epoch": 2.48,
+      "learning_rate": 1.4463573983949341e-05,
+      "loss": 1.5502,
+      "step": 310
+    },
+    {
+      "epoch": 2.52,
+      "learning_rate": 1.2369331995613665e-05,
+      "loss": 1.4882,
+      "step": 315
+    },
+    {
+      "epoch": 2.56,
+      "learning_rate": 1.042882397605871e-05,
+      "loss": 1.6666,
+      "step": 320
+    },
+    {
+      "epoch": 2.6,
+      "learning_rate": 8.645454235739903e-06,
+      "loss": 1.4874,
+      "step": 325
+    },
+    {
+      "epoch": 2.64,
+      "learning_rate": 7.022351411174866e-06,
+      "loss": 1.6158,
+      "step": 330
+    },
+    {
+      "epoch": 2.68,
+      "learning_rate": 5.562362976251901e-06,
+      "loss": 1.4376,
+      "step": 335
+    },
+    {
+      "epoch": 2.72,
+      "learning_rate": 4.268050246793276e-06,
+      "loss": 1.6202,
+      "step": 340
+    },
+    {
+      "epoch": 2.76,
+      "learning_rate": 3.1416838871368924e-06,
+      "loss": 1.5493,
+      "step": 345
+    },
+    {
+      "epoch": 2.8,
+      "learning_rate": 2.1852399266194314e-06,
+      "loss": 1.6157,
+      "step": 350
+    },
+    {
+      "epoch": 2.84,
+      "learning_rate": 1.400396292949513e-06,
+      "loss": 1.5631,
+      "step": 355
+    },
+    {
+      "epoch": 2.88,
+      "learning_rate": 7.885298685522235e-07,
+      "loss": 1.5326,
+      "step": 360
+    },
+    {
+      "epoch": 2.92,
+      "learning_rate": 3.50714075049563e-07,
+      "loss": 1.5978,
+      "step": 365
+    },
+    {
+      "epoch": 2.96,
+      "learning_rate": 8.771699011416168e-08,
+      "loss": 1.4251,
+      "step": 370
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 0.0,
+      "loss": 1.6276,
+      "step": 375
+    },
+    {
+      "epoch": 3.0,
+      "step": 375,
+      "total_flos": 4.32034079145984e+16,
+      "train_loss": 1.628477378845215,
+      "train_runtime": 1045.5144,
+      "train_samples_per_second": 5.739,
+      "train_steps_per_second": 0.359
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 375,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 100,
+  "total_flos": 4.32034079145984e+16,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:84990e28d454e290f6393201e221ba051e01af18e581a4f5994ac8396ad7c48b
+size 4920