l40 p2 completed 6 epochs

llama-factory/saves/internlm2_5_7b/lora/sft_bf16_p2_full/README.md

@@ -0,0 +1,72 @@
+---
+license: other
+library_name: peft
+tags:
+- llama-factory
+- lora
+- generated_from_trainer
+base_model: internlm/internlm2_5-7b-chat-1m
+metrics:
+- accuracy
+model-index:
+- name: sft_bf16_p2_full
+  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. -->
+
+# sft_bf16_p2_full
+
+This model is a fine-tuned version of [internlm/internlm2_5-7b-chat-1m](https://huggingface.co/internlm/internlm2_5-7b-chat-1m) on the alpaca_mgtv_p2 dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.4369
+- Accuracy: 0.8984
+
+## 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: 0.0001
+- train_batch_size: 32
+- eval_batch_size: 1
+- seed: 42
+- gradient_accumulation_steps: 8
+- total_train_batch_size: 256
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_ratio: 0.1
+- num_epochs: 6.0
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss | Accuracy |
+|:-------------:|:-----:|:----:|:---------------:|:--------:|
+| 0.277         | 1.0   | 88   | 0.2600          | 0.9031   |
+| 0.2427        | 2.0   | 176  | 0.2574          | 0.9006   |
+| 0.1813        | 3.0   | 264  | 0.2706          | 0.9027   |
+| 0.1263        | 4.0   | 352  | 0.2946          | 0.8994   |
+| 0.0684        | 5.0   | 440  | 0.3700          | 0.8997   |
+| 0.0486        | 6.0   | 528  | 0.4369          | 0.8984   |
+
+
+### Framework versions
+
+- PEFT 0.11.1
+- Transformers 4.41.2
+- Pytorch 2.2.1+cu121
+- Datasets 2.19.1
+- Tokenizers 0.19.1

llama-factory/saves/internlm2_5_7b/lora/sft_bf16_p2_full/adapter_config.json

@@ -0,0 +1,32 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "internlm/internlm2_5-7b-chat-1m",
+  "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": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 8,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "wqkv",
+    "w2",
+    "w1",
+    "w3",
+    "wo"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}

llama-factory/saves/internlm2_5_7b/lora/sft_bf16_p2_full/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:92a2e3824954685b0ea8fb8a04cc2b4db2cfebb48037425fd844501b0c301050
+size 75539712

llama-factory/saves/internlm2_5_7b/lora/sft_bf16_p2_full/all_results.json

@@ -0,0 +1,13 @@
+{
+    "epoch": 6.0,
+    "eval_accuracy": 0.8984,
+    "eval_loss": 0.4368518590927124,
+    "eval_runtime": 232.12,
+    "eval_samples_per_second": 10.77,
+    "eval_steps_per_second": 10.77,
+    "total_flos": 2.9041541335076045e+18,
+    "train_loss": 0.28717788867652416,
+    "train_runtime": 47077.7992,
+    "train_samples_per_second": 2.868,
+    "train_steps_per_second": 0.011
+}

llama-factory/saves/internlm2_5_7b/lora/sft_bf16_p2_full/checkpoint-440/README.md

@@ -0,0 +1,202 @@
+---
+library_name: peft
+base_model: internlm/internlm2_5-7b-chat-1m
+---
+
+# 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 0.11.1

llama-factory/saves/internlm2_5_7b/lora/sft_bf16_p2_full/checkpoint-440/adapter_config.json

@@ -0,0 +1,32 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "internlm/internlm2_5-7b-chat-1m",
+  "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": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 8,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "wqkv",
+    "w2",
+    "w1",
+    "w3",
+    "wo"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}

llama-factory/saves/internlm2_5_7b/lora/sft_bf16_p2_full/checkpoint-440/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:628a2f898fdb7dab1660bfaa12309eff585e40d4c692fffe6795adc705dcea80
+size 75539712

llama-factory/saves/internlm2_5_7b/lora/sft_bf16_p2_full/checkpoint-440/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6fda0cb1eed9e9d314b3d0f546b49de8616dcb126afb643e9cb8f07e3a0df7ad
+size 151264058

llama-factory/saves/internlm2_5_7b/lora/sft_bf16_p2_full/checkpoint-440/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d245e05e72192c132e0f2edb6fdcae0c578c890f0fe912f17ec7b0bba2d38cc3
+size 14244

llama-factory/saves/internlm2_5_7b/lora/sft_bf16_p2_full/checkpoint-440/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bbca99b341582421991e0d53c661a42ff44b04d0e45e04965ef5d464774e7259
+size 1064

llama-factory/saves/internlm2_5_7b/lora/sft_bf16_p2_full/checkpoint-440/special_tokens_map.json

@@ -0,0 +1,38 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|action_start|>",
+    "<|action_end|>",
+    "<|interpreter|>",
+    "<|plugin|>"
+  ],
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

llama-factory/saves/internlm2_5_7b/lora/sft_bf16_p2_full/checkpoint-440/tokenization_internlm2.py

@@ -0,0 +1,236 @@
+# coding=utf-8
+# Copyright (c) The InternLM team and The HuggingFace Inc. team. All rights reserved.
+#
+# This code is based on transformers/src/transformers/models/llama/tokenization_llama.py
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+"""Tokenization classes for InternLM."""
+import os
+from shutil import copyfile
+from typing import Any, Dict, List, Optional, Tuple
+
+import sentencepiece as spm
+from transformers.tokenization_utils import PreTrainedTokenizer
+from transformers.utils import logging
+
+logger = logging.get_logger(__name__)
+
+VOCAB_FILES_NAMES = {"vocab_file": "./tokenizer.model"}
+
+PRETRAINED_VOCAB_FILES_MAP = {}
+
+
+# Modified from transformers.model.llama.tokenization_llama.LlamaTokenizer
+class InternLM2Tokenizer(PreTrainedTokenizer):
+    """
+    Construct a InternLM2 tokenizer. Based on byte-level Byte-Pair-Encoding.
+
+    Args:
+        vocab_file (`str`):
+            Path to the vocabulary file.
+    """
+
+    vocab_files_names = VOCAB_FILES_NAMES
+    pretrained_vocab_files_map = PRETRAINED_VOCAB_FILES_MAP
+    model_input_names = ["input_ids", "attention_mask"]
+    _auto_class = "AutoTokenizer"
+
+    def __init__(
+        self,
+        vocab_file,
+        unk_token="<unk>",
+        bos_token="<s>",
+        eos_token="</s>",
+        pad_token="</s>",
+        sp_model_kwargs: Optional[Dict[str, Any]] = None,
+        add_bos_token=True,
+        add_eos_token=False,
+        decode_with_prefix_space=False,
+        clean_up_tokenization_spaces=False,
+        **kwargs,
+    ):
+        self.sp_model_kwargs = {} if sp_model_kwargs is None else sp_model_kwargs
+        self.vocab_file = vocab_file
+        self.add_bos_token = add_bos_token
+        self.add_eos_token = add_eos_token
+        self.decode_with_prefix_space = decode_with_prefix_space
+        self.sp_model = spm.SentencePieceProcessor(**self.sp_model_kwargs)
+        self.sp_model.Load(vocab_file)
+        self._no_prefix_space_tokens = None
+        super().__init__(
+            bos_token=bos_token,
+            eos_token=eos_token,
+            unk_token=unk_token,
+            pad_token=pad_token,
+            clean_up_tokenization_spaces=clean_up_tokenization_spaces,
+            **kwargs,
+        )
+
+    @property
+    def no_prefix_space_tokens(self):
+        if self._no_prefix_space_tokens is None:
+            vocab = self.convert_ids_to_tokens(list(range(self.vocab_size)))
+            self._no_prefix_space_tokens = {i for i, tok in enumerate(vocab) if not tok.startswith("▁")}
+        return self._no_prefix_space_tokens
+
+    @property
+    def vocab_size(self):
+        """Returns vocab size"""
+        return self.sp_model.get_piece_size()
+
+    @property
+    def bos_token_id(self) -> Optional[int]:
+        return self.sp_model.bos_id()
+
+    @property
+    def eos_token_id(self) -> Optional[int]:
+        return self.sp_model.eos_id()
+
+    def get_vocab(self):
+        """Returns vocab as a dict"""
+        vocab = {self.convert_ids_to_tokens(i): i for i in range(self.vocab_size)}
+        vocab.update(self.added_tokens_encoder)
+        return vocab
+
+    def _tokenize(self, text):
+        """Returns a tokenized string."""
+        return self.sp_model.encode(text, out_type=str)
+
+    def _convert_token_to_id(self, token):
+        """Converts a token (str) in an id using the vocab."""
+        return self.sp_model.piece_to_id(token)
+
+    def _convert_id_to_token(self, index):
+        """Converts an index (integer) in a token (str) using the vocab."""
+        token = self.sp_model.IdToPiece(index)
+        return token
+
+    def _maybe_add_prefix_space(self, tokens, decoded):
+        if tokens and tokens[0] not in self.no_prefix_space_tokens:
+            return " " + decoded
+        else:
+            return decoded
+
+    def convert_tokens_to_string(self, tokens):
+        """Converts a sequence of tokens (string) in a single string."""
+        current_sub_tokens = []
+        out_string = ""
+        prev_is_special = False
+        for token in tokens:
+            # make sure that special tokens are not decoded using sentencepiece model
+            if token in self.all_special_tokens:
+                if not prev_is_special:
+                    out_string += " "
+                out_string += self.sp_model.decode(current_sub_tokens) + token
+                prev_is_special = True
+                current_sub_tokens = []
+            else:
+                current_sub_tokens.append(token)
+                prev_is_special = False
+        out_string += self.sp_model.decode(current_sub_tokens)
+        out_string = self.clean_up_tokenization(out_string)
+        out_string = self._maybe_add_prefix_space(tokens=tokens, decoded=out_string)
+        return out_string[1:]
+
+    def save_vocabulary(self, save_directory, filename_prefix: Optional[str] = None) -> Tuple[str]:
+        """
+        Save the vocabulary and special tokens file to a directory.
+
+        Args:
+            save_directory (`str`):
+                The directory in which to save the vocabulary.
+
+        Returns:
+            `Tuple(str)`: Paths to the files saved.
+        """
+        if not os.path.isdir(save_directory):
+            logger.error(f"Vocabulary path ({save_directory}) should be a directory")
+            return
+        out_vocab_file = os.path.join(
+            save_directory, (filename_prefix + "-" if filename_prefix else "") + VOCAB_FILES_NAMES["vocab_file"]
+        )
+
+        if os.path.abspath(self.vocab_file) != os.path.abspath(out_vocab_file) and os.path.isfile(self.vocab_file):
+            copyfile(self.vocab_file, out_vocab_file)
+        elif not os.path.isfile(self.vocab_file):
+            with open(out_vocab_file, "wb") as fi:
+                content_spiece_model = self.sp_model.serialized_model_proto()
+                fi.write(content_spiece_model)
+
+        return (out_vocab_file,)
+
+    def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=None):
+        if self.add_bos_token:
+            bos_token_ids = [self.bos_token_id]
+        else:
+            bos_token_ids = []
+
+        output = bos_token_ids + token_ids_0
+
+        if token_ids_1 is not None:
+            output = output + token_ids_1
+
+        if self.add_eos_token:
+            output = output + [self.eos_token_id]
+
+        return output
+
+    def get_special_tokens_mask(
+        self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None, already_has_special_tokens: bool = False
+    ) -> List[int]:
+        """
+        Retrieve sequence ids from a token list that has no special tokens added. This method is called when adding
+        special tokens using the tokenizer `prepare_for_model` method.
+
+        Args:
+            token_ids_0 (`List[int]`):
+                List of IDs.
+            token_ids_1 (`List[int]`, *optional*):
+                Optional second list of IDs for sequence pairs.
+            already_has_special_tokens (`bool`, *optional*, defaults to `False`):
+                Whether or not the token list is already formatted with special tokens for the model.
+
+        Returns:
+            `List[int]`: A list of integers in the range [0, 1]: 1 for a special token, 0 for a sequence token.
+        """
+        if already_has_special_tokens:
+            return super().get_special_tokens_mask(
+                token_ids_0=token_ids_0, token_ids_1=token_ids_1, already_has_special_tokens=True
+            )
+
+        if token_ids_1 is None:
+            return [1] + ([0] * len(token_ids_0)) + [1]
+        return [1] + ([0] * len(token_ids_0)) + [1, 1] + ([0] * len(token_ids_1)) + [1]
+
+    def create_token_type_ids_from_sequences(
+        self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None
+    ) -> List[int]:
+        """
+        Create a mask from the two sequences passed to be used in a sequence-pair classification task. T5 does not make
+        use of token type ids, therefore a list of zeros is returned.
+
+        Args:
+            token_ids_0 (`List[int]`):
+                List of IDs.
+            token_ids_1 (`List[int]`, *optional*):
+                Optional second list of IDs for sequence pairs.
+
+        Returns:
+            `List[int]`: List of zeros.
+        """
+        eos = [self.eos_token_id]
+
+        if token_ids_1 is None:
+            return len(token_ids_0 + eos) * [0]
+        return len(token_ids_0 + eos + token_ids_1 + eos) * [0]

llama-factory/saves/internlm2_5_7b/lora/sft_bf16_p2_full/checkpoint-440/tokenization_internlm2_fast.py

@@ -0,0 +1,214 @@
+# coding=utf-8
+# Copyright (c) The InternLM team and The HuggingFace Inc. team. All rights reserved.
+#
+# This code is based on transformers/src/transformers/models/llama/tokenization_llama_fast.py
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+"""Tokenization Fast class for InternLM."""
+import os
+from shutil import copyfile
+from typing import Any, Dict, Optional, Tuple
+
+from tokenizers import processors, decoders, Tokenizer, normalizers
+from tokenizers.models import BPE
+
+from transformers.tokenization_utils_fast import PreTrainedTokenizerFast
+from transformers.utils import logging
+
+from transformers.convert_slow_tokenizer import (
+    SLOW_TO_FAST_CONVERTERS,
+    SpmConverter,
+    SentencePieceExtractor,
+)
+
+from .tokenization_internlm2 import InternLM2Tokenizer
+
+logger = logging.get_logger(__name__)
+
+VOCAB_FILES_NAMES = {"vocab_file": "./tokenizer.model"}
+
+# Modified from transformers.convert_slow_tokenizer.LlamaConverter
+class InternLM2Converter(SpmConverter):
+    handle_byte_fallback = True
+
+    def vocab(self, proto):
+        vocab = [
+            ("<unk>", 0.0),
+            ("<s>", 0.0),
+            ("</s>", 0.0),
+        ]
+        vocab += [(piece.piece, piece.score) for piece in proto.pieces[3:]]
+        return vocab
+
+    def unk_id(self, proto):
+        unk_id = 0
+        return unk_id
+
+    def decoder(self, replacement, add_prefix_space):
+        decoders_sequence = [
+            decoders.Replace("▁", " "),
+            decoders.ByteFallback(),
+            decoders.Fuse(),
+        ]
+        if self.proto.normalizer_spec.add_dummy_prefix:
+            decoders_sequence.append(decoders.Strip(content=" ", left=1))
+        return decoders.Sequence(decoders_sequence)
+
+    def tokenizer(self, proto):
+        model_type = proto.trainer_spec.model_type
+        vocab_scores = self.vocab(proto)
+        # special tokens
+        added_tokens = self.original_tokenizer.added_tokens_decoder
+        for i in range(len(vocab_scores)):
+            piece, score = vocab_scores[i]
+            if i in added_tokens:
+                vocab_scores[i] = (added_tokens[i].content, score)
+        if model_type == 1:
+            raise RuntimeError("InternLM2 is supposed to be a BPE model!")
+
+        elif model_type == 2:
+            _, merges = SentencePieceExtractor(self.original_tokenizer.vocab_file).extract(vocab_scores)
+            bpe_vocab = {word: i for i, (word, _score) in enumerate(vocab_scores)}
+            tokenizer = Tokenizer(
+                BPE(bpe_vocab, merges, unk_token=proto.trainer_spec.unk_piece, fuse_unk=True, byte_fallback=True)
+            )
+            tokenizer.add_special_tokens(
+                [ added_token for index, added_token in added_tokens.items()]
+            )
+        else:
+            raise Exception(
+                "You're trying to run a `Unigram` model but you're file was trained with a different algorithm"
+            )
+
+        return tokenizer
+
+    def normalizer(self, proto):
+        normalizers_list = []
+        if proto.normalizer_spec.add_dummy_prefix:
+            normalizers_list.append(normalizers.Prepend(prepend="▁"))
+        normalizers_list.append(normalizers.Replace(pattern=" ", content="▁"))
+        return normalizers.Sequence(normalizers_list)
+
+    def pre_tokenizer(self, replacement, add_prefix_space):
+        return None
+
+SLOW_TO_FAST_CONVERTERS["InternLM2Tokenizer"] = InternLM2Converter
+
+
+# Modified from transformers.model.llama.tokenization_llama_fast.LlamaTokenizerFast -> InternLM2TokenizerFast
+class InternLM2TokenizerFast(PreTrainedTokenizerFast):
+    vocab_files_names = VOCAB_FILES_NAMES
+    slow_tokenizer_class = InternLM2Tokenizer
+    padding_side = "left"
+    model_input_names = ["input_ids", "attention_mask"]
+    _auto_class = "AutoTokenizer"
+
+    def __init__(
+        self,
+        vocab_file,
+        unk_token="<unk>",
+        bos_token="<s>",
+        eos_token="</s>",
+        pad_token="</s>",
+        sp_model_kwargs: Optional[Dict[str, Any]] = None,
+        add_bos_token=True,
+        add_eos_token=False,
+        decode_with_prefix_space=False,
+        clean_up_tokenization_spaces=False,
+        **kwargs,
+    ):
+        super().__init__(
+            vocab_file=vocab_file,
+            unk_token=unk_token,
+            bos_token=bos_token,
+            eos_token=eos_token,
+            pad_token=pad_token,
+            sp_model_kwargs=sp_model_kwargs,
+            add_bos_token=add_bos_token,
+            add_eos_token=add_eos_token,
+            decode_with_prefix_space=decode_with_prefix_space,
+            clean_up_tokenization_spaces=clean_up_tokenization_spaces,
+            **kwargs,
+        )
+        self._add_bos_token = add_bos_token
+        self._add_eos_token = add_eos_token
+        self.update_post_processor()
+        self.vocab_file = vocab_file
+
+    @property
+    def can_save_slow_tokenizer(self) -> bool:
+        return os.path.isfile(self.vocab_file) if self.vocab_file else False
+
+    def update_post_processor(self):
+        """
+        Updates the underlying post processor with the current `bos_token` and `eos_token`.
+        """
+        bos = self.bos_token
+        bos_token_id = self.bos_token_id
+        if bos is None and self.add_bos_token:
+            raise ValueError("add_bos_token = True but bos_token = None")
+
+        eos = self.eos_token
+        eos_token_id = self.eos_token_id
+        if eos is None and self.add_eos_token:
+            raise ValueError("add_eos_token = True but eos_token = None")
+
+        single = f"{(bos+':0 ') if self.add_bos_token else ''}$A:0{(' '+eos+':0') if self.add_eos_token else ''}"
+        pair = f"{single}{(' '+bos+':1') if self.add_bos_token else ''} $B:1{(' '+eos+':1') if self.add_eos_token else ''}"
+
+        special_tokens = []
+        if self.add_bos_token:
+            special_tokens.append((bos, bos_token_id))
+        if self.add_eos_token:
+            special_tokens.append((eos, eos_token_id))
+        self._tokenizer.post_processor = processors.TemplateProcessing(
+            single=single, pair=pair, special_tokens=special_tokens
+        )
+
+    @property
+    def add_eos_token(self):
+        return self._add_eos_token
+
+    @property
+    def add_bos_token(self):
+        return self._add_bos_token
+
+    @add_eos_token.setter
+    def add_eos_token(self, value):
+        self._add_eos_token = value
+        self.update_post_processor()
+
+    @add_bos_token.setter
+    def add_bos_token(self, value):
+        self._add_bos_token = value
+        self.update_post_processor()
+
+    def save_vocabulary(self, save_directory: str, filename_prefix: Optional[str] = None) -> Tuple[str]:
+        if not self.can_save_slow_tokenizer:
+            raise ValueError(
+                "Your fast tokenizer does not have the necessary information to save the vocabulary for a slow "
+                "tokenizer."
+            )
+
+        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):
+            copyfile(self.vocab_file, out_vocab_file)
+
+        return (out_vocab_file,)

llama-factory/saves/internlm2_5_7b/lora/sft_bf16_p2_full/checkpoint-440/tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f868398fc4e05ee1e8aeba95ddf18ddcc45b8bce55d5093bead5bbf80429b48b
+size 1477754

llama-factory/saves/internlm2_5_7b/lora/sft_bf16_p2_full/checkpoint-440/tokenizer_config.json

@@ -0,0 +1,104 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "92538": {
+      "content": "<|plugin|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "92539": {
+      "content": "<|interpreter|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "92540": {
+      "content": "<|action_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "92541": {
+      "content": "<|action_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "92542": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "92543": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|action_start|>",
+    "<|action_end|>",
+    "<|interpreter|>",
+    "<|plugin|>"
+  ],
+  "auto_map": {
+    "AutoTokenizer": [
+      "tokenization_internlm2.InternLM2Tokenizer",
+      "tokenization_internlm2_fast.InternLM2TokenizerFast"
+    ]
+  },
+  "bos_token": "<s>",
+  "chat_template": "{{ '<s>' }}{% if messages[0]['role'] == 'system' %}{% set system_message = messages[0]['content'] %}{% endif %}{% if system_message is defined %}{{ '<|im_start|>system\n' + system_message + '<|im_end|>\n' }}{% endif %}{% for message in messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ '<|im_start|>user\n' + content + '<|im_end|>\n<|im_start|>assistant\n' }}{% elif message['role'] == 'assistant' %}{{ content + '<|im_end|>\n' }}{% endif %}{% endfor %}",
+  "clean_up_tokenization_spaces": false,
+  "decode_with_prefix_space": false,
+  "eos_token": "</s>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "</s>",
+  "padding_side": "right",
+  "sp_model_kwargs": null,
+  "split_special_tokens": false,
+  "tokenizer_class": "InternLM2Tokenizer",
+  "unk_token": "<unk>"
+}

llama-factory/saves/internlm2_5_7b/lora/sft_bf16_p2_full/checkpoint-440/trainer_state.json

@@ -0,0 +1,386 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 5.0,
+  "eval_steps": 88,
+  "global_step": 440,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.11363636363636363,
+      "grad_norm": 3.514838457107544,
+      "learning_rate": 1.8867924528301888e-05,
+      "loss": 6.943,
+      "step": 10
+    },
+    {
+      "epoch": 0.22727272727272727,
+      "grad_norm": 1.0595427751541138,
+      "learning_rate": 3.7735849056603776e-05,
+      "loss": 0.446,
+      "step": 20
+    },
+    {
+      "epoch": 0.3409090909090909,
+      "grad_norm": 0.6256385445594788,
+      "learning_rate": 5.660377358490566e-05,
+      "loss": 0.3515,
+      "step": 30
+    },
+    {
+      "epoch": 0.45454545454545453,
+      "grad_norm": 0.633573055267334,
+      "learning_rate": 7.547169811320755e-05,
+      "loss": 0.288,
+      "step": 40
+    },
+    {
+      "epoch": 0.5681818181818182,
+      "grad_norm": 0.4915701746940613,
+      "learning_rate": 9.433962264150944e-05,
+      "loss": 0.2819,
+      "step": 50
+    },
+    {
+      "epoch": 0.6818181818181818,
+      "grad_norm": 0.40083640813827515,
+      "learning_rate": 9.994642390694308e-05,
+      "loss": 0.2765,
+      "step": 60
+    },
+    {
+      "epoch": 0.7954545454545454,
+      "grad_norm": 0.7176418304443359,
+      "learning_rate": 9.968428675226714e-05,
+      "loss": 0.2754,
+      "step": 70
+    },
+    {
+      "epoch": 0.9090909090909091,
+      "grad_norm": 0.6853049397468567,
+      "learning_rate": 9.92048928531717e-05,
+      "loss": 0.277,
+      "step": 80
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.9030666666666668,
+      "eval_loss": 0.26003387570381165,
+      "eval_runtime": 231.9406,
+      "eval_samples_per_second": 10.779,
+      "eval_steps_per_second": 10.779,
+      "step": 88
+    },
+    {
+      "epoch": 1.0227272727272727,
+      "grad_norm": 0.4268323481082916,
+      "learning_rate": 9.851033847720166e-05,
+      "loss": 0.2619,
+      "step": 90
+    },
+    {
+      "epoch": 1.1363636363636362,
+      "grad_norm": 0.9503114819526672,
+      "learning_rate": 9.760366073392246e-05,
+      "loss": 0.2385,
+      "step": 100
+    },
+    {
+      "epoch": 1.25,
+      "grad_norm": 0.3606574237346649,
+      "learning_rate": 9.648882429441257e-05,
+      "loss": 0.2341,
+      "step": 110
+    },
+    {
+      "epoch": 1.3636363636363638,
+      "grad_norm": 0.7226484417915344,
+      "learning_rate": 9.517070405476575e-05,
+      "loss": 0.2447,
+      "step": 120
+    },
+    {
+      "epoch": 1.4772727272727273,
+      "grad_norm": 0.8543397188186646,
+      "learning_rate": 9.365506381941066e-05,
+      "loss": 0.2441,
+      "step": 130
+    },
+    {
+      "epoch": 1.5909090909090908,
+      "grad_norm": 0.800394594669342,
+      "learning_rate": 9.194853109746074e-05,
+      "loss": 0.2379,
+      "step": 140
+    },
+    {
+      "epoch": 1.7045454545454546,
+      "grad_norm": 0.5756838321685791,
+      "learning_rate": 9.005856812230304e-05,
+      "loss": 0.2434,
+      "step": 150
+    },
+    {
+      "epoch": 1.8181818181818183,
+      "grad_norm": 1.0771032571792603,
+      "learning_rate": 8.799343922115044e-05,
+      "loss": 0.2352,
+      "step": 160
+    },
+    {
+      "epoch": 1.9318181818181817,
+      "grad_norm": 0.4805872440338135,
+      "learning_rate": 8.576217467724128e-05,
+      "loss": 0.2427,
+      "step": 170
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.9006,
+      "eval_loss": 0.25738978385925293,
+      "eval_runtime": 231.2131,
+      "eval_samples_per_second": 10.813,
+      "eval_steps_per_second": 10.813,
+      "step": 176
+    },
+    {
+      "epoch": 2.0454545454545454,
+      "grad_norm": 0.5219587683677673,
+      "learning_rate": 8.337453124270863e-05,
+      "loss": 0.22,
+      "step": 180
+    },
+    {
+      "epoch": 2.159090909090909,
+      "grad_norm": 0.6363154053688049,
+      "learning_rate": 8.084094947478556e-05,
+      "loss": 0.1787,
+      "step": 190
+    },
+    {
+      "epoch": 2.2727272727272725,
+      "grad_norm": 0.6807820796966553,
+      "learning_rate": 7.817250808190483e-05,
+      "loss": 0.1647,
+      "step": 200
+    },
+    {
+      "epoch": 2.3863636363636362,
+      "grad_norm": 0.5443515777587891,
+      "learning_rate": 7.538087547932585e-05,
+      "loss": 0.1828,
+      "step": 210
+    },
+    {
+      "epoch": 2.5,
+      "grad_norm": 0.4641902446746826,
+      "learning_rate": 7.247825876612353e-05,
+      "loss": 0.1782,
+      "step": 220
+    },
+    {
+      "epoch": 2.6136363636363638,
+      "grad_norm": 0.5865933299064636,
+      "learning_rate": 6.947735034665002e-05,
+      "loss": 0.1942,
+      "step": 230
+    },
+    {
+      "epoch": 2.7272727272727275,
+      "grad_norm": 0.5332173705101013,
+      "learning_rate": 6.639127242987988e-05,
+      "loss": 0.1852,
+      "step": 240
+    },
+    {
+      "epoch": 2.840909090909091,
+      "grad_norm": 0.5550218820571899,
+      "learning_rate": 6.323351964932908e-05,
+      "loss": 0.1936,
+      "step": 250
+    },
+    {
+      "epoch": 2.9545454545454546,
+      "grad_norm": 0.6850063800811768,
+      "learning_rate": 6.001790005445607e-05,
+      "loss": 0.1813,
+      "step": 260
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.9027,
+      "eval_loss": 0.2705931067466736,
+      "eval_runtime": 231.0113,
+      "eval_samples_per_second": 10.822,
+      "eval_steps_per_second": 10.822,
+      "step": 264
+    },
+    {
+      "epoch": 3.0681818181818183,
+      "grad_norm": 0.42733630537986755,
+      "learning_rate": 5.675847473157485e-05,
+      "loss": 0.14,
+      "step": 270
+    },
+    {
+      "epoch": 3.1818181818181817,
+      "grad_norm": 0.5972977876663208,
+      "learning_rate": 5.3469496318302204e-05,
+      "loss": 0.1197,
+      "step": 280
+    },
+    {
+      "epoch": 3.2954545454545454,
+      "grad_norm": 0.4995785653591156,
+      "learning_rate": 5.016534668039976e-05,
+      "loss": 0.1198,
+      "step": 290
+    },
+    {
+      "epoch": 3.409090909090909,
+      "grad_norm": 0.5500032305717468,
+      "learning_rate": 4.6860474023534335e-05,
+      "loss": 0.1131,
+      "step": 300
+    },
+    {
+      "epoch": 3.5227272727272725,
+      "grad_norm": 0.4452584683895111,
+      "learning_rate": 4.3569329714950704e-05,
+      "loss": 0.1185,
+      "step": 310
+    },
+    {
+      "epoch": 3.6363636363636362,
+      "grad_norm": 0.4754205346107483,
+      "learning_rate": 4.0306305091319595e-05,
+      "loss": 0.1197,
+      "step": 320
+    },
+    {
+      "epoch": 3.75,
+      "grad_norm": 0.6347799301147461,
+      "learning_rate": 3.7085668529084184e-05,
+      "loss": 0.122,
+      "step": 330
+    },
+    {
+      "epoch": 3.8636363636363638,
+      "grad_norm": 0.48911160230636597,
+      "learning_rate": 3.392150305248024e-05,
+      "loss": 0.1163,
+      "step": 340
+    },
+    {
+      "epoch": 3.9772727272727275,
+      "grad_norm": 0.6460514068603516,
+      "learning_rate": 3.082764475205442e-05,
+      "loss": 0.1263,
+      "step": 350
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.8993666666666668,
+      "eval_loss": 0.2945823669433594,
+      "eval_runtime": 233.0443,
+      "eval_samples_per_second": 10.728,
+      "eval_steps_per_second": 10.728,
+      "step": 352
+    },
+    {
+      "epoch": 4.090909090909091,
+      "grad_norm": 0.37293320894241333,
+      "learning_rate": 2.7817622282960815e-05,
+      "loss": 0.0856,
+      "step": 360
+    },
+    {
+      "epoch": 4.204545454545454,
+      "grad_norm": 0.5676562190055847,
+      "learning_rate": 2.490459770759398e-05,
+      "loss": 0.0661,
+      "step": 370
+    },
+    {
+      "epoch": 4.318181818181818,
+      "grad_norm": 0.5680781006813049,
+      "learning_rate": 2.2101308941239203e-05,
+      "loss": 0.061,
+      "step": 380
+    },
+    {
+      "epoch": 4.431818181818182,
+      "grad_norm": 0.690169095993042,
+      "learning_rate": 1.942001405240979e-05,
+      "loss": 0.0744,
+      "step": 390
+    },
+    {
+      "epoch": 4.545454545454545,
+      "grad_norm": 0.5858839750289917,
+      "learning_rate": 1.6872437661432517e-05,
+      "loss": 0.0736,
+      "step": 400
+    },
+    {
+      "epoch": 4.659090909090909,
+      "grad_norm": 0.6473811268806458,
+      "learning_rate": 1.4469719671666043e-05,
+      "loss": 0.0779,
+      "step": 410
+    },
+    {
+      "epoch": 4.7727272727272725,
+      "grad_norm": 0.3694300055503845,
+      "learning_rate": 1.2222366557537911e-05,
+      "loss": 0.075,
+      "step": 420
+    },
+    {
+      "epoch": 4.886363636363637,
+      "grad_norm": 0.5935441851615906,
+      "learning_rate": 1.0140205422405214e-05,
+      "loss": 0.0752,
+      "step": 430
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 0.7272607684135437,
+      "learning_rate": 8.232341027131885e-06,
+      "loss": 0.0684,
+      "step": 440
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.8996666666666668,
+      "eval_loss": 0.369967520236969,
+      "eval_runtime": 232.4252,
+      "eval_samples_per_second": 10.756,
+      "eval_steps_per_second": 10.756,
+      "step": 440
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 528,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 6,
+  "save_steps": 88,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 2.420046524577743e+18,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}

llama-factory/saves/internlm2_5_7b/lora/sft_bf16_p2_full/checkpoint-440/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9d5c2bfb768f1720511bef6e983b38242156f823ff83caba648f7520ad9d664c
+size 5304

llama-factory/saves/internlm2_5_7b/lora/sft_bf16_p2_full/checkpoint-528/README.md

@@ -0,0 +1,202 @@
+---
+library_name: peft
+base_model: internlm/internlm2_5-7b-chat-1m
+---
+
+# 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 0.11.1

llama-factory/saves/internlm2_5_7b/lora/sft_bf16_p2_full/checkpoint-528/adapter_config.json

@@ -0,0 +1,32 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "internlm/internlm2_5-7b-chat-1m",
+  "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": 0.0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 8,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "wqkv",
+    "w2",
+    "w1",
+    "w3",
+    "wo"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}

llama-factory/saves/internlm2_5_7b/lora/sft_bf16_p2_full/checkpoint-528/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:92a2e3824954685b0ea8fb8a04cc2b4db2cfebb48037425fd844501b0c301050
+size 75539712

llama-factory/saves/internlm2_5_7b/lora/sft_bf16_p2_full/checkpoint-528/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a24c92a3e00afe3d9c670aa108bdfe4207517a2fec62d3e953239e6b69ff592d
+size 151264058

llama-factory/saves/internlm2_5_7b/lora/sft_bf16_p2_full/checkpoint-528/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e3e5d946241df2516b06d7074d8779088eae7607173ad780df56583910a9589b
+size 14244

llama-factory/saves/internlm2_5_7b/lora/sft_bf16_p2_full/checkpoint-528/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f760752c4266a113f3e31ae4fd0ef06b844dcb204fbf0c40e6c00f707b3df523
+size 1064

llama-factory/saves/internlm2_5_7b/lora/sft_bf16_p2_full/checkpoint-528/special_tokens_map.json

@@ -0,0 +1,38 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|action_start|>",
+    "<|action_end|>",
+    "<|interpreter|>",
+    "<|plugin|>"
+  ],
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

llama-factory/saves/internlm2_5_7b/lora/sft_bf16_p2_full/checkpoint-528/tokenization_internlm2.py

@@ -0,0 +1,236 @@
+# coding=utf-8
+# Copyright (c) The InternLM team and The HuggingFace Inc. team. All rights reserved.
+#
+# This code is based on transformers/src/transformers/models/llama/tokenization_llama.py
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+"""Tokenization classes for InternLM."""
+import os
+from shutil import copyfile
+from typing import Any, Dict, List, Optional, Tuple
+
+import sentencepiece as spm
+from transformers.tokenization_utils import PreTrainedTokenizer
+from transformers.utils import logging
+
+logger = logging.get_logger(__name__)
+
+VOCAB_FILES_NAMES = {"vocab_file": "./tokenizer.model"}
+
+PRETRAINED_VOCAB_FILES_MAP = {}
+
+
+# Modified from transformers.model.llama.tokenization_llama.LlamaTokenizer
+class InternLM2Tokenizer(PreTrainedTokenizer):
+    """
+    Construct a InternLM2 tokenizer. Based on byte-level Byte-Pair-Encoding.
+
+    Args:
+        vocab_file (`str`):
+            Path to the vocabulary file.
+    """
+
+    vocab_files_names = VOCAB_FILES_NAMES
+    pretrained_vocab_files_map = PRETRAINED_VOCAB_FILES_MAP
+    model_input_names = ["input_ids", "attention_mask"]
+    _auto_class = "AutoTokenizer"
+
+    def __init__(
+        self,
+        vocab_file,
+        unk_token="<unk>",
+        bos_token="<s>",
+        eos_token="</s>",
+        pad_token="</s>",
+        sp_model_kwargs: Optional[Dict[str, Any]] = None,
+        add_bos_token=True,
+        add_eos_token=False,
+        decode_with_prefix_space=False,
+        clean_up_tokenization_spaces=False,
+        **kwargs,
+    ):
+        self.sp_model_kwargs = {} if sp_model_kwargs is None else sp_model_kwargs
+        self.vocab_file = vocab_file
+        self.add_bos_token = add_bos_token
+        self.add_eos_token = add_eos_token
+        self.decode_with_prefix_space = decode_with_prefix_space
+        self.sp_model = spm.SentencePieceProcessor(**self.sp_model_kwargs)
+        self.sp_model.Load(vocab_file)
+        self._no_prefix_space_tokens = None
+        super().__init__(
+            bos_token=bos_token,
+            eos_token=eos_token,
+            unk_token=unk_token,
+            pad_token=pad_token,
+            clean_up_tokenization_spaces=clean_up_tokenization_spaces,
+            **kwargs,
+        )
+
+    @property
+    def no_prefix_space_tokens(self):
+        if self._no_prefix_space_tokens is None:
+            vocab = self.convert_ids_to_tokens(list(range(self.vocab_size)))
+            self._no_prefix_space_tokens = {i for i, tok in enumerate(vocab) if not tok.startswith("▁")}
+        return self._no_prefix_space_tokens
+
+    @property
+    def vocab_size(self):
+        """Returns vocab size"""
+        return self.sp_model.get_piece_size()
+
+    @property
+    def bos_token_id(self) -> Optional[int]:
+        return self.sp_model.bos_id()
+
+    @property
+    def eos_token_id(self) -> Optional[int]:
+        return self.sp_model.eos_id()
+
+    def get_vocab(self):
+        """Returns vocab as a dict"""
+        vocab = {self.convert_ids_to_tokens(i): i for i in range(self.vocab_size)}
+        vocab.update(self.added_tokens_encoder)
+        return vocab
+
+    def _tokenize(self, text):
+        """Returns a tokenized string."""
+        return self.sp_model.encode(text, out_type=str)
+
+    def _convert_token_to_id(self, token):
+        """Converts a token (str) in an id using the vocab."""
+        return self.sp_model.piece_to_id(token)
+
+    def _convert_id_to_token(self, index):
+        """Converts an index (integer) in a token (str) using the vocab."""
+        token = self.sp_model.IdToPiece(index)
+        return token
+
+    def _maybe_add_prefix_space(self, tokens, decoded):
+        if tokens and tokens[0] not in self.no_prefix_space_tokens:
+            return " " + decoded
+        else:
+            return decoded
+
+    def convert_tokens_to_string(self, tokens):
+        """Converts a sequence of tokens (string) in a single string."""
+        current_sub_tokens = []
+        out_string = ""
+        prev_is_special = False
+        for token in tokens:
+            # make sure that special tokens are not decoded using sentencepiece model
+            if token in self.all_special_tokens:
+                if not prev_is_special:
+                    out_string += " "
+                out_string += self.sp_model.decode(current_sub_tokens) + token
+                prev_is_special = True
+                current_sub_tokens = []
+            else:
+                current_sub_tokens.append(token)
+                prev_is_special = False
+        out_string += self.sp_model.decode(current_sub_tokens)
+        out_string = self.clean_up_tokenization(out_string)
+        out_string = self._maybe_add_prefix_space(tokens=tokens, decoded=out_string)
+        return out_string[1:]
+
+    def save_vocabulary(self, save_directory, filename_prefix: Optional[str] = None) -> Tuple[str]:
+        """
+        Save the vocabulary and special tokens file to a directory.
+
+        Args:
+            save_directory (`str`):
+                The directory in which to save the vocabulary.
+
+        Returns:
+            `Tuple(str)`: Paths to the files saved.
+        """
+        if not os.path.isdir(save_directory):
+            logger.error(f"Vocabulary path ({save_directory}) should be a directory")
+            return
+        out_vocab_file = os.path.join(
+            save_directory, (filename_prefix + "-" if filename_prefix else "") + VOCAB_FILES_NAMES["vocab_file"]
+        )
+
+        if os.path.abspath(self.vocab_file) != os.path.abspath(out_vocab_file) and os.path.isfile(self.vocab_file):
+            copyfile(self.vocab_file, out_vocab_file)
+        elif not os.path.isfile(self.vocab_file):
+            with open(out_vocab_file, "wb") as fi:
+                content_spiece_model = self.sp_model.serialized_model_proto()
+                fi.write(content_spiece_model)
+
+        return (out_vocab_file,)
+
+    def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=None):
+        if self.add_bos_token:
+            bos_token_ids = [self.bos_token_id]
+        else:
+            bos_token_ids = []
+
+        output = bos_token_ids + token_ids_0
+
+        if token_ids_1 is not None:
+            output = output + token_ids_1
+
+        if self.add_eos_token:
+            output = output + [self.eos_token_id]
+
+        return output
+
+    def get_special_tokens_mask(
+        self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None, already_has_special_tokens: bool = False
+    ) -> List[int]:
+        """
+        Retrieve sequence ids from a token list that has no special tokens added. This method is called when adding
+        special tokens using the tokenizer `prepare_for_model` method.
+
+        Args:
+            token_ids_0 (`List[int]`):
+                List of IDs.
+            token_ids_1 (`List[int]`, *optional*):
+                Optional second list of IDs for sequence pairs.
+            already_has_special_tokens (`bool`, *optional*, defaults to `False`):
+                Whether or not the token list is already formatted with special tokens for the model.
+
+        Returns:
+            `List[int]`: A list of integers in the range [0, 1]: 1 for a special token, 0 for a sequence token.
+        """
+        if already_has_special_tokens:
+            return super().get_special_tokens_mask(
+                token_ids_0=token_ids_0, token_ids_1=token_ids_1, already_has_special_tokens=True
+            )
+
+        if token_ids_1 is None:
+            return [1] + ([0] * len(token_ids_0)) + [1]
+        return [1] + ([0] * len(token_ids_0)) + [1, 1] + ([0] * len(token_ids_1)) + [1]
+
+    def create_token_type_ids_from_sequences(
+        self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None
+    ) -> List[int]:
+        """
+        Create a mask from the two sequences passed to be used in a sequence-pair classification task. T5 does not make
+        use of token type ids, therefore a list of zeros is returned.
+
+        Args:
+            token_ids_0 (`List[int]`):
+                List of IDs.
+            token_ids_1 (`List[int]`, *optional*):
+                Optional second list of IDs for sequence pairs.
+
+        Returns:
+            `List[int]`: List of zeros.
+        """
+        eos = [self.eos_token_id]
+
+        if token_ids_1 is None:
+            return len(token_ids_0 + eos) * [0]
+        return len(token_ids_0 + eos + token_ids_1 + eos) * [0]

llama-factory/saves/internlm2_5_7b/lora/sft_bf16_p2_full/checkpoint-528/tokenization_internlm2_fast.py

@@ -0,0 +1,214 @@
+# coding=utf-8
+# Copyright (c) The InternLM team and The HuggingFace Inc. team. All rights reserved.
+#
+# This code is based on transformers/src/transformers/models/llama/tokenization_llama_fast.py
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+"""Tokenization Fast class for InternLM."""
+import os
+from shutil import copyfile
+from typing import Any, Dict, Optional, Tuple
+
+from tokenizers import processors, decoders, Tokenizer, normalizers
+from tokenizers.models import BPE
+
+from transformers.tokenization_utils_fast import PreTrainedTokenizerFast
+from transformers.utils import logging
+
+from transformers.convert_slow_tokenizer import (
+    SLOW_TO_FAST_CONVERTERS,
+    SpmConverter,
+    SentencePieceExtractor,
+)
+
+from .tokenization_internlm2 import InternLM2Tokenizer
+
+logger = logging.get_logger(__name__)
+
+VOCAB_FILES_NAMES = {"vocab_file": "./tokenizer.model"}
+
+# Modified from transformers.convert_slow_tokenizer.LlamaConverter
+class InternLM2Converter(SpmConverter):
+    handle_byte_fallback = True
+
+    def vocab(self, proto):
+        vocab = [
+            ("<unk>", 0.0),
+            ("<s>", 0.0),
+            ("</s>", 0.0),
+        ]
+        vocab += [(piece.piece, piece.score) for piece in proto.pieces[3:]]
+        return vocab
+
+    def unk_id(self, proto):
+        unk_id = 0
+        return unk_id
+
+    def decoder(self, replacement, add_prefix_space):
+        decoders_sequence = [
+            decoders.Replace("▁", " "),
+            decoders.ByteFallback(),
+            decoders.Fuse(),
+        ]
+        if self.proto.normalizer_spec.add_dummy_prefix:
+            decoders_sequence.append(decoders.Strip(content=" ", left=1))
+        return decoders.Sequence(decoders_sequence)
+
+    def tokenizer(self, proto):
+        model_type = proto.trainer_spec.model_type
+        vocab_scores = self.vocab(proto)
+        # special tokens
+        added_tokens = self.original_tokenizer.added_tokens_decoder
+        for i in range(len(vocab_scores)):
+            piece, score = vocab_scores[i]
+            if i in added_tokens:
+                vocab_scores[i] = (added_tokens[i].content, score)
+        if model_type == 1:
+            raise RuntimeError("InternLM2 is supposed to be a BPE model!")
+
+        elif model_type == 2:
+            _, merges = SentencePieceExtractor(self.original_tokenizer.vocab_file).extract(vocab_scores)
+            bpe_vocab = {word: i for i, (word, _score) in enumerate(vocab_scores)}
+            tokenizer = Tokenizer(
+                BPE(bpe_vocab, merges, unk_token=proto.trainer_spec.unk_piece, fuse_unk=True, byte_fallback=True)
+            )
+            tokenizer.add_special_tokens(
+                [ added_token for index, added_token in added_tokens.items()]
+            )
+        else:
+            raise Exception(
+                "You're trying to run a `Unigram` model but you're file was trained with a different algorithm"
+            )
+
+        return tokenizer
+
+    def normalizer(self, proto):
+        normalizers_list = []
+        if proto.normalizer_spec.add_dummy_prefix:
+            normalizers_list.append(normalizers.Prepend(prepend="▁"))
+        normalizers_list.append(normalizers.Replace(pattern=" ", content="▁"))
+        return normalizers.Sequence(normalizers_list)
+
+    def pre_tokenizer(self, replacement, add_prefix_space):
+        return None
+
+SLOW_TO_FAST_CONVERTERS["InternLM2Tokenizer"] = InternLM2Converter
+
+
+# Modified from transformers.model.llama.tokenization_llama_fast.LlamaTokenizerFast -> InternLM2TokenizerFast
+class InternLM2TokenizerFast(PreTrainedTokenizerFast):
+    vocab_files_names = VOCAB_FILES_NAMES
+    slow_tokenizer_class = InternLM2Tokenizer
+    padding_side = "left"
+    model_input_names = ["input_ids", "attention_mask"]
+    _auto_class = "AutoTokenizer"
+
+    def __init__(
+        self,
+        vocab_file,
+        unk_token="<unk>",
+        bos_token="<s>",
+        eos_token="</s>",
+        pad_token="</s>",
+        sp_model_kwargs: Optional[Dict[str, Any]] = None,
+        add_bos_token=True,
+        add_eos_token=False,
+        decode_with_prefix_space=False,
+        clean_up_tokenization_spaces=False,
+        **kwargs,
+    ):
+        super().__init__(
+            vocab_file=vocab_file,
+            unk_token=unk_token,
+            bos_token=bos_token,
+            eos_token=eos_token,
+            pad_token=pad_token,
+            sp_model_kwargs=sp_model_kwargs,
+            add_bos_token=add_bos_token,
+            add_eos_token=add_eos_token,
+            decode_with_prefix_space=decode_with_prefix_space,
+            clean_up_tokenization_spaces=clean_up_tokenization_spaces,
+            **kwargs,
+        )
+        self._add_bos_token = add_bos_token
+        self._add_eos_token = add_eos_token
+        self.update_post_processor()
+        self.vocab_file = vocab_file
+
+    @property
+    def can_save_slow_tokenizer(self) -> bool:
+        return os.path.isfile(self.vocab_file) if self.vocab_file else False
+
+    def update_post_processor(self):
+        """
+        Updates the underlying post processor with the current `bos_token` and `eos_token`.
+        """
+        bos = self.bos_token
+        bos_token_id = self.bos_token_id
+        if bos is None and self.add_bos_token:
+            raise ValueError("add_bos_token = True but bos_token = None")
+
+        eos = self.eos_token
+        eos_token_id = self.eos_token_id
+        if eos is None and self.add_eos_token:
+            raise ValueError("add_eos_token = True but eos_token = None")
+
+        single = f"{(bos+':0 ') if self.add_bos_token else ''}$A:0{(' '+eos+':0') if self.add_eos_token else ''}"
+        pair = f"{single}{(' '+bos+':1') if self.add_bos_token else ''} $B:1{(' '+eos+':1') if self.add_eos_token else ''}"
+
+        special_tokens = []
+        if self.add_bos_token:
+            special_tokens.append((bos, bos_token_id))
+        if self.add_eos_token:
+            special_tokens.append((eos, eos_token_id))
+        self._tokenizer.post_processor = processors.TemplateProcessing(
+            single=single, pair=pair, special_tokens=special_tokens
+        )
+
+    @property
+    def add_eos_token(self):
+        return self._add_eos_token
+
+    @property
+    def add_bos_token(self):
+        return self._add_bos_token
+
+    @add_eos_token.setter
+    def add_eos_token(self, value):
+        self._add_eos_token = value
+        self.update_post_processor()
+
+    @add_bos_token.setter
+    def add_bos_token(self, value):
+        self._add_bos_token = value
+        self.update_post_processor()
+
+    def save_vocabulary(self, save_directory: str, filename_prefix: Optional[str] = None) -> Tuple[str]:
+        if not self.can_save_slow_tokenizer:
+            raise ValueError(
+                "Your fast tokenizer does not have the necessary information to save the vocabulary for a slow "
+                "tokenizer."
+            )
+
+        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):
+            copyfile(self.vocab_file, out_vocab_file)
+
+        return (out_vocab_file,)

llama-factory/saves/internlm2_5_7b/lora/sft_bf16_p2_full/checkpoint-528/tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f868398fc4e05ee1e8aeba95ddf18ddcc45b8bce55d5093bead5bbf80429b48b
+size 1477754

llama-factory/saves/internlm2_5_7b/lora/sft_bf16_p2_full/checkpoint-528/tokenizer_config.json

@@ -0,0 +1,104 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "92538": {
+      "content": "<|plugin|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "92539": {
+      "content": "<|interpreter|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "92540": {
+      "content": "<|action_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "92541": {
+      "content": "<|action_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "92542": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "92543": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|action_start|>",
+    "<|action_end|>",
+    "<|interpreter|>",
+    "<|plugin|>"
+  ],
+  "auto_map": {
+    "AutoTokenizer": [
+      "tokenization_internlm2.InternLM2Tokenizer",
+      "tokenization_internlm2_fast.InternLM2TokenizerFast"
+    ]
+  },
+  "bos_token": "<s>",
+  "chat_template": "{{ '<s>' }}{% if messages[0]['role'] == 'system' %}{% set system_message = messages[0]['content'] %}{% endif %}{% if system_message is defined %}{{ '<|im_start|>system\n' + system_message + '<|im_end|>\n' }}{% endif %}{% for message in messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ '<|im_start|>user\n' + content + '<|im_end|>\n<|im_start|>assistant\n' }}{% elif message['role'] == 'assistant' %}{{ content + '<|im_end|>\n' }}{% endif %}{% endfor %}",
+  "clean_up_tokenization_spaces": false,
+  "decode_with_prefix_space": false,
+  "eos_token": "</s>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "</s>",
+  "padding_side": "right",
+  "sp_model_kwargs": null,
+  "split_special_tokens": false,
+  "tokenizer_class": "InternLM2Tokenizer",
+  "unk_token": "<unk>"
+}

llama-factory/saves/internlm2_5_7b/lora/sft_bf16_p2_full/checkpoint-528/trainer_state.json

@@ -0,0 +1,451 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 6.0,
+  "eval_steps": 88,
+  "global_step": 528,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.11363636363636363,
+      "grad_norm": 3.514838457107544,
+      "learning_rate": 1.8867924528301888e-05,
+      "loss": 6.943,
+      "step": 10
+    },
+    {
+      "epoch": 0.22727272727272727,
+      "grad_norm": 1.0595427751541138,
+      "learning_rate": 3.7735849056603776e-05,
+      "loss": 0.446,
+      "step": 20
+    },
+    {
+      "epoch": 0.3409090909090909,
+      "grad_norm": 0.6256385445594788,
+      "learning_rate": 5.660377358490566e-05,
+      "loss": 0.3515,
+      "step": 30
+    },
+    {
+      "epoch": 0.45454545454545453,
+      "grad_norm": 0.633573055267334,
+      "learning_rate": 7.547169811320755e-05,
+      "loss": 0.288,
+      "step": 40
+    },
+    {
+      "epoch": 0.5681818181818182,
+      "grad_norm": 0.4915701746940613,
+      "learning_rate": 9.433962264150944e-05,
+      "loss": 0.2819,
+      "step": 50
+    },
+    {
+      "epoch": 0.6818181818181818,
+      "grad_norm": 0.40083640813827515,
+      "learning_rate": 9.994642390694308e-05,
+      "loss": 0.2765,
+      "step": 60
+    },
+    {
+      "epoch": 0.7954545454545454,
+      "grad_norm": 0.7176418304443359,
+      "learning_rate": 9.968428675226714e-05,
+      "loss": 0.2754,
+      "step": 70
+    },
+    {
+      "epoch": 0.9090909090909091,
+      "grad_norm": 0.6853049397468567,
+      "learning_rate": 9.92048928531717e-05,
+      "loss": 0.277,
+      "step": 80
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.9030666666666668,
+      "eval_loss": 0.26003387570381165,
+      "eval_runtime": 231.9406,
+      "eval_samples_per_second": 10.779,
+      "eval_steps_per_second": 10.779,
+      "step": 88
+    },
+    {
+      "epoch": 1.0227272727272727,
+      "grad_norm": 0.4268323481082916,
+      "learning_rate": 9.851033847720166e-05,
+      "loss": 0.2619,
+      "step": 90
+    },
+    {
+      "epoch": 1.1363636363636362,
+      "grad_norm": 0.9503114819526672,
+      "learning_rate": 9.760366073392246e-05,
+      "loss": 0.2385,
+      "step": 100
+    },
+    {
+      "epoch": 1.25,
+      "grad_norm": 0.3606574237346649,
+      "learning_rate": 9.648882429441257e-05,
+      "loss": 0.2341,
+      "step": 110
+    },
+    {
+      "epoch": 1.3636363636363638,
+      "grad_norm": 0.7226484417915344,
+      "learning_rate": 9.517070405476575e-05,
+      "loss": 0.2447,
+      "step": 120
+    },
+    {
+      "epoch": 1.4772727272727273,
+      "grad_norm": 0.8543397188186646,
+      "learning_rate": 9.365506381941066e-05,
+      "loss": 0.2441,
+      "step": 130
+    },
+    {
+      "epoch": 1.5909090909090908,
+      "grad_norm": 0.800394594669342,
+      "learning_rate": 9.194853109746074e-05,
+      "loss": 0.2379,
+      "step": 140
+    },
+    {
+      "epoch": 1.7045454545454546,
+      "grad_norm": 0.5756838321685791,
+      "learning_rate": 9.005856812230304e-05,
+      "loss": 0.2434,
+      "step": 150
+    },
+    {
+      "epoch": 1.8181818181818183,
+      "grad_norm": 1.0771032571792603,
+      "learning_rate": 8.799343922115044e-05,
+      "loss": 0.2352,
+      "step": 160
+    },
+    {
+      "epoch": 1.9318181818181817,
+      "grad_norm": 0.4805872440338135,
+      "learning_rate": 8.576217467724128e-05,
+      "loss": 0.2427,
+      "step": 170
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.9006,
+      "eval_loss": 0.25738978385925293,
+      "eval_runtime": 231.2131,
+      "eval_samples_per_second": 10.813,
+      "eval_steps_per_second": 10.813,
+      "step": 176
+    },
+    {
+      "epoch": 2.0454545454545454,
+      "grad_norm": 0.5219587683677673,
+      "learning_rate": 8.337453124270863e-05,
+      "loss": 0.22,
+      "step": 180
+    },
+    {
+      "epoch": 2.159090909090909,
+      "grad_norm": 0.6363154053688049,
+      "learning_rate": 8.084094947478556e-05,
+      "loss": 0.1787,
+      "step": 190
+    },
+    {
+      "epoch": 2.2727272727272725,
+      "grad_norm": 0.6807820796966553,
+      "learning_rate": 7.817250808190483e-05,
+      "loss": 0.1647,
+      "step": 200
+    },
+    {
+      "epoch": 2.3863636363636362,
+      "grad_norm": 0.5443515777587891,
+      "learning_rate": 7.538087547932585e-05,
+      "loss": 0.1828,
+      "step": 210
+    },
+    {
+      "epoch": 2.5,
+      "grad_norm": 0.4641902446746826,
+      "learning_rate": 7.247825876612353e-05,
+      "loss": 0.1782,
+      "step": 220
+    },
+    {
+      "epoch": 2.6136363636363638,
+      "grad_norm": 0.5865933299064636,
+      "learning_rate": 6.947735034665002e-05,
+      "loss": 0.1942,
+      "step": 230
+    },
+    {
+      "epoch": 2.7272727272727275,
+      "grad_norm": 0.5332173705101013,
+      "learning_rate": 6.639127242987988e-05,
+      "loss": 0.1852,
+      "step": 240
+    },
+    {
+      "epoch": 2.840909090909091,
+      "grad_norm": 0.5550218820571899,
+      "learning_rate": 6.323351964932908e-05,
+      "loss": 0.1936,
+      "step": 250
+    },
+    {
+      "epoch": 2.9545454545454546,
+      "grad_norm": 0.6850063800811768,
+      "learning_rate": 6.001790005445607e-05,
+      "loss": 0.1813,
+      "step": 260
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.9027,
+      "eval_loss": 0.2705931067466736,
+      "eval_runtime": 231.0113,
+      "eval_samples_per_second": 10.822,
+      "eval_steps_per_second": 10.822,
+      "step": 264
+    },
+    {
+      "epoch": 3.0681818181818183,
+      "grad_norm": 0.42733630537986755,
+      "learning_rate": 5.675847473157485e-05,
+      "loss": 0.14,
+      "step": 270
+    },
+    {
+      "epoch": 3.1818181818181817,
+      "grad_norm": 0.5972977876663208,
+      "learning_rate": 5.3469496318302204e-05,
+      "loss": 0.1197,
+      "step": 280
+    },
+    {
+      "epoch": 3.2954545454545454,
+      "grad_norm": 0.4995785653591156,
+      "learning_rate": 5.016534668039976e-05,
+      "loss": 0.1198,
+      "step": 290
+    },
+    {
+      "epoch": 3.409090909090909,
+      "grad_norm": 0.5500032305717468,
+      "learning_rate": 4.6860474023534335e-05,
+      "loss": 0.1131,
+      "step": 300
+    },
+    {
+      "epoch": 3.5227272727272725,
+      "grad_norm": 0.4452584683895111,
+      "learning_rate": 4.3569329714950704e-05,
+      "loss": 0.1185,
+      "step": 310
+    },
+    {
+      "epoch": 3.6363636363636362,
+      "grad_norm": 0.4754205346107483,
+      "learning_rate": 4.0306305091319595e-05,
+      "loss": 0.1197,
+      "step": 320
+    },
+    {
+      "epoch": 3.75,
+      "grad_norm": 0.6347799301147461,
+      "learning_rate": 3.7085668529084184e-05,
+      "loss": 0.122,
+      "step": 330
+    },
+    {
+      "epoch": 3.8636363636363638,
+      "grad_norm": 0.48911160230636597,
+      "learning_rate": 3.392150305248024e-05,
+      "loss": 0.1163,
+      "step": 340
+    },
+    {
+      "epoch": 3.9772727272727275,
+      "grad_norm": 0.6460514068603516,
+      "learning_rate": 3.082764475205442e-05,
+      "loss": 0.1263,
+      "step": 350
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.8993666666666668,
+      "eval_loss": 0.2945823669433594,
+      "eval_runtime": 233.0443,
+      "eval_samples_per_second": 10.728,
+      "eval_steps_per_second": 10.728,
+      "step": 352
+    },
+    {
+      "epoch": 4.090909090909091,
+      "grad_norm": 0.37293320894241333,
+      "learning_rate": 2.7817622282960815e-05,
+      "loss": 0.0856,
+      "step": 360
+    },
+    {
+      "epoch": 4.204545454545454,
+      "grad_norm": 0.5676562190055847,
+      "learning_rate": 2.490459770759398e-05,
+      "loss": 0.0661,
+      "step": 370
+    },
+    {
+      "epoch": 4.318181818181818,
+      "grad_norm": 0.5680781006813049,
+      "learning_rate": 2.2101308941239203e-05,
+      "loss": 0.061,
+      "step": 380
+    },
+    {
+      "epoch": 4.431818181818182,
+      "grad_norm": 0.690169095993042,
+      "learning_rate": 1.942001405240979e-05,
+      "loss": 0.0744,
+      "step": 390
+    },
+    {
+      "epoch": 4.545454545454545,
+      "grad_norm": 0.5858839750289917,
+      "learning_rate": 1.6872437661432517e-05,
+      "loss": 0.0736,
+      "step": 400
+    },
+    {
+      "epoch": 4.659090909090909,
+      "grad_norm": 0.6473811268806458,
+      "learning_rate": 1.4469719671666043e-05,
+      "loss": 0.0779,
+      "step": 410
+    },
+    {
+      "epoch": 4.7727272727272725,
+      "grad_norm": 0.3694300055503845,
+      "learning_rate": 1.2222366557537911e-05,
+      "loss": 0.075,
+      "step": 420
+    },
+    {
+      "epoch": 4.886363636363637,
+      "grad_norm": 0.5935441851615906,
+      "learning_rate": 1.0140205422405214e-05,
+      "loss": 0.0752,
+      "step": 430
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 0.7272607684135437,
+      "learning_rate": 8.232341027131885e-06,
+      "loss": 0.0684,
+      "step": 440
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.8996666666666668,
+      "eval_loss": 0.369967520236969,
+      "eval_runtime": 232.4252,
+      "eval_samples_per_second": 10.756,
+      "eval_steps_per_second": 10.756,
+      "step": 440
+    },
+    {
+      "epoch": 5.113636363636363,
+      "grad_norm": 0.3855780363082886,
+      "learning_rate": 6.5071159772861436e-06,
+      "loss": 0.0531,
+      "step": 450
+    },
+    {
+      "epoch": 5.2272727272727275,
+      "grad_norm": 0.3242223560810089,
+      "learning_rate": 4.972074243048897e-06,
+      "loss": 0.0437,
+      "step": 460
+    },
+    {
+      "epoch": 5.340909090909091,
+      "grad_norm": 0.36955130100250244,
+      "learning_rate": 3.6339281713517303e-06,
+      "loss": 0.0463,
+      "step": 470
+    },
+    {
+      "epoch": 5.454545454545454,
+      "grad_norm": 0.3851165473461151,
+      "learning_rate": 2.4985291344915674e-06,
+      "loss": 0.0485,
+      "step": 480
+    },
+    {
+      "epoch": 5.568181818181818,
+      "grad_norm": 0.30520951747894287,
+      "learning_rate": 1.5708419435684462e-06,
+      "loss": 0.0495,
+      "step": 490
+    },
+    {
+      "epoch": 5.681818181818182,
+      "grad_norm": 0.8094011545181274,
+      "learning_rate": 8.549231386298151e-07,
+      "loss": 0.0484,
+      "step": 500
+    },
+    {
+      "epoch": 5.795454545454545,
+      "grad_norm": 0.21888971328735352,
+      "learning_rate": 3.5390325045304706e-07,
+      "loss": 0.0384,
+      "step": 510
+    },
+    {
+      "epoch": 5.909090909090909,
+      "grad_norm": 0.4017506539821625,
+      "learning_rate": 6.997311153086883e-08,
+      "loss": 0.0486,
+      "step": 520
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.8984,
+      "eval_loss": 0.4368518590927124,
+      "eval_runtime": 232.9443,
+      "eval_samples_per_second": 10.732,
+      "eval_steps_per_second": 10.732,
+      "step": 528
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 528,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 6,
+  "save_steps": 88,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 2.9041541335076045e+18,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}

llama-factory/saves/internlm2_5_7b/lora/sft_bf16_p2_full/checkpoint-528/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9d5c2bfb768f1720511bef6e983b38242156f823ff83caba648f7520ad9d664c
+size 5304

llama-factory/saves/internlm2_5_7b/lora/sft_bf16_p2_full/eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 6.0,
+    "eval_accuracy": 0.8984,
+    "eval_loss": 0.4368518590927124,
+    "eval_runtime": 232.12,
+    "eval_samples_per_second": 10.77,
+    "eval_steps_per_second": 10.77
+}

llama-factory/saves/internlm2_5_7b/lora/sft_bf16_p2_full/special_tokens_map.json

@@ -0,0 +1,38 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|action_start|>",
+    "<|action_end|>",
+    "<|interpreter|>",
+    "<|plugin|>"
+  ],
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

llama-factory/saves/internlm2_5_7b/lora/sft_bf16_p2_full/tokenization_internlm2.py

@@ -0,0 +1,236 @@
+# coding=utf-8
+# Copyright (c) The InternLM team and The HuggingFace Inc. team. All rights reserved.
+#
+# This code is based on transformers/src/transformers/models/llama/tokenization_llama.py
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+"""Tokenization classes for InternLM."""
+import os
+from shutil import copyfile
+from typing import Any, Dict, List, Optional, Tuple
+
+import sentencepiece as spm
+from transformers.tokenization_utils import PreTrainedTokenizer
+from transformers.utils import logging
+
+logger = logging.get_logger(__name__)
+
+VOCAB_FILES_NAMES = {"vocab_file": "./tokenizer.model"}
+
+PRETRAINED_VOCAB_FILES_MAP = {}
+
+
+# Modified from transformers.model.llama.tokenization_llama.LlamaTokenizer
+class InternLM2Tokenizer(PreTrainedTokenizer):
+    """
+    Construct a InternLM2 tokenizer. Based on byte-level Byte-Pair-Encoding.
+
+    Args:
+        vocab_file (`str`):
+            Path to the vocabulary file.
+    """
+
+    vocab_files_names = VOCAB_FILES_NAMES
+    pretrained_vocab_files_map = PRETRAINED_VOCAB_FILES_MAP
+    model_input_names = ["input_ids", "attention_mask"]
+    _auto_class = "AutoTokenizer"
+
+    def __init__(
+        self,
+        vocab_file,
+        unk_token="<unk>",
+        bos_token="<s>",
+        eos_token="</s>",
+        pad_token="</s>",
+        sp_model_kwargs: Optional[Dict[str, Any]] = None,
+        add_bos_token=True,
+        add_eos_token=False,
+        decode_with_prefix_space=False,
+        clean_up_tokenization_spaces=False,
+        **kwargs,
+    ):
+        self.sp_model_kwargs = {} if sp_model_kwargs is None else sp_model_kwargs
+        self.vocab_file = vocab_file
+        self.add_bos_token = add_bos_token
+        self.add_eos_token = add_eos_token
+        self.decode_with_prefix_space = decode_with_prefix_space
+        self.sp_model = spm.SentencePieceProcessor(**self.sp_model_kwargs)
+        self.sp_model.Load(vocab_file)
+        self._no_prefix_space_tokens = None
+        super().__init__(
+            bos_token=bos_token,
+            eos_token=eos_token,
+            unk_token=unk_token,
+            pad_token=pad_token,
+            clean_up_tokenization_spaces=clean_up_tokenization_spaces,
+            **kwargs,
+        )
+
+    @property
+    def no_prefix_space_tokens(self):
+        if self._no_prefix_space_tokens is None:
+            vocab = self.convert_ids_to_tokens(list(range(self.vocab_size)))
+            self._no_prefix_space_tokens = {i for i, tok in enumerate(vocab) if not tok.startswith("▁")}
+        return self._no_prefix_space_tokens
+
+    @property
+    def vocab_size(self):
+        """Returns vocab size"""
+        return self.sp_model.get_piece_size()
+
+    @property
+    def bos_token_id(self) -> Optional[int]:
+        return self.sp_model.bos_id()
+
+    @property
+    def eos_token_id(self) -> Optional[int]:
+        return self.sp_model.eos_id()
+
+    def get_vocab(self):
+        """Returns vocab as a dict"""
+        vocab = {self.convert_ids_to_tokens(i): i for i in range(self.vocab_size)}
+        vocab.update(self.added_tokens_encoder)
+        return vocab
+
+    def _tokenize(self, text):
+        """Returns a tokenized string."""
+        return self.sp_model.encode(text, out_type=str)
+
+    def _convert_token_to_id(self, token):
+        """Converts a token (str) in an id using the vocab."""
+        return self.sp_model.piece_to_id(token)
+
+    def _convert_id_to_token(self, index):
+        """Converts an index (integer) in a token (str) using the vocab."""
+        token = self.sp_model.IdToPiece(index)
+        return token
+
+    def _maybe_add_prefix_space(self, tokens, decoded):
+        if tokens and tokens[0] not in self.no_prefix_space_tokens:
+            return " " + decoded
+        else:
+            return decoded
+
+    def convert_tokens_to_string(self, tokens):
+        """Converts a sequence of tokens (string) in a single string."""
+        current_sub_tokens = []
+        out_string = ""
+        prev_is_special = False
+        for token in tokens:
+            # make sure that special tokens are not decoded using sentencepiece model
+            if token in self.all_special_tokens:
+                if not prev_is_special:
+                    out_string += " "
+                out_string += self.sp_model.decode(current_sub_tokens) + token
+                prev_is_special = True
+                current_sub_tokens = []
+            else:
+                current_sub_tokens.append(token)
+                prev_is_special = False
+        out_string += self.sp_model.decode(current_sub_tokens)
+        out_string = self.clean_up_tokenization(out_string)
+        out_string = self._maybe_add_prefix_space(tokens=tokens, decoded=out_string)
+        return out_string[1:]
+
+    def save_vocabulary(self, save_directory, filename_prefix: Optional[str] = None) -> Tuple[str]:
+        """
+        Save the vocabulary and special tokens file to a directory.
+
+        Args:
+            save_directory (`str`):
+                The directory in which to save the vocabulary.
+
+        Returns:
+            `Tuple(str)`: Paths to the files saved.
+        """
+        if not os.path.isdir(save_directory):
+            logger.error(f"Vocabulary path ({save_directory}) should be a directory")
+            return
+        out_vocab_file = os.path.join(
+            save_directory, (filename_prefix + "-" if filename_prefix else "") + VOCAB_FILES_NAMES["vocab_file"]
+        )
+
+        if os.path.abspath(self.vocab_file) != os.path.abspath(out_vocab_file) and os.path.isfile(self.vocab_file):
+            copyfile(self.vocab_file, out_vocab_file)
+        elif not os.path.isfile(self.vocab_file):
+            with open(out_vocab_file, "wb") as fi:
+                content_spiece_model = self.sp_model.serialized_model_proto()
+                fi.write(content_spiece_model)
+
+        return (out_vocab_file,)
+
+    def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=None):
+        if self.add_bos_token:
+            bos_token_ids = [self.bos_token_id]
+        else:
+            bos_token_ids = []
+
+        output = bos_token_ids + token_ids_0
+
+        if token_ids_1 is not None:
+            output = output + token_ids_1
+
+        if self.add_eos_token:
+            output = output + [self.eos_token_id]
+
+        return output
+
+    def get_special_tokens_mask(
+        self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None, already_has_special_tokens: bool = False
+    ) -> List[int]:
+        """
+        Retrieve sequence ids from a token list that has no special tokens added. This method is called when adding
+        special tokens using the tokenizer `prepare_for_model` method.
+
+        Args:
+            token_ids_0 (`List[int]`):
+                List of IDs.
+            token_ids_1 (`List[int]`, *optional*):
+                Optional second list of IDs for sequence pairs.
+            already_has_special_tokens (`bool`, *optional*, defaults to `False`):
+                Whether or not the token list is already formatted with special tokens for the model.
+
+        Returns:
+            `List[int]`: A list of integers in the range [0, 1]: 1 for a special token, 0 for a sequence token.
+        """
+        if already_has_special_tokens:
+            return super().get_special_tokens_mask(
+                token_ids_0=token_ids_0, token_ids_1=token_ids_1, already_has_special_tokens=True
+            )
+
+        if token_ids_1 is None:
+            return [1] + ([0] * len(token_ids_0)) + [1]
+        return [1] + ([0] * len(token_ids_0)) + [1, 1] + ([0] * len(token_ids_1)) + [1]
+
+    def create_token_type_ids_from_sequences(
+        self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None
+    ) -> List[int]:
+        """
+        Create a mask from the two sequences passed to be used in a sequence-pair classification task. T5 does not make
+        use of token type ids, therefore a list of zeros is returned.
+
+        Args:
+            token_ids_0 (`List[int]`):
+                List of IDs.
+            token_ids_1 (`List[int]`, *optional*):
+                Optional second list of IDs for sequence pairs.
+
+        Returns:
+            `List[int]`: List of zeros.
+        """
+        eos = [self.eos_token_id]
+
+        if token_ids_1 is None:
+            return len(token_ids_0 + eos) * [0]
+        return len(token_ids_0 + eos + token_ids_1 + eos) * [0]

llama-factory/saves/internlm2_5_7b/lora/sft_bf16_p2_full/tokenization_internlm2_fast.py

@@ -0,0 +1,214 @@
+# coding=utf-8
+# Copyright (c) The InternLM team and The HuggingFace Inc. team. All rights reserved.
+#
+# This code is based on transformers/src/transformers/models/llama/tokenization_llama_fast.py
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+"""Tokenization Fast class for InternLM."""
+import os
+from shutil import copyfile
+from typing import Any, Dict, Optional, Tuple
+
+from tokenizers import processors, decoders, Tokenizer, normalizers
+from tokenizers.models import BPE
+
+from transformers.tokenization_utils_fast import PreTrainedTokenizerFast
+from transformers.utils import logging
+
+from transformers.convert_slow_tokenizer import (
+    SLOW_TO_FAST_CONVERTERS,
+    SpmConverter,
+    SentencePieceExtractor,
+)
+
+from .tokenization_internlm2 import InternLM2Tokenizer
+
+logger = logging.get_logger(__name__)
+
+VOCAB_FILES_NAMES = {"vocab_file": "./tokenizer.model"}
+
+# Modified from transformers.convert_slow_tokenizer.LlamaConverter
+class InternLM2Converter(SpmConverter):
+    handle_byte_fallback = True
+
+    def vocab(self, proto):
+        vocab = [
+            ("<unk>", 0.0),
+            ("<s>", 0.0),
+            ("</s>", 0.0),
+        ]
+        vocab += [(piece.piece, piece.score) for piece in proto.pieces[3:]]
+        return vocab
+
+    def unk_id(self, proto):
+        unk_id = 0
+        return unk_id
+
+    def decoder(self, replacement, add_prefix_space):
+        decoders_sequence = [
+            decoders.Replace("▁", " "),
+            decoders.ByteFallback(),
+            decoders.Fuse(),
+        ]
+        if self.proto.normalizer_spec.add_dummy_prefix:
+            decoders_sequence.append(decoders.Strip(content=" ", left=1))
+        return decoders.Sequence(decoders_sequence)
+
+    def tokenizer(self, proto):
+        model_type = proto.trainer_spec.model_type
+        vocab_scores = self.vocab(proto)
+        # special tokens
+        added_tokens = self.original_tokenizer.added_tokens_decoder
+        for i in range(len(vocab_scores)):
+            piece, score = vocab_scores[i]
+            if i in added_tokens:
+                vocab_scores[i] = (added_tokens[i].content, score)
+        if model_type == 1:
+            raise RuntimeError("InternLM2 is supposed to be a BPE model!")
+
+        elif model_type == 2:
+            _, merges = SentencePieceExtractor(self.original_tokenizer.vocab_file).extract(vocab_scores)
+            bpe_vocab = {word: i for i, (word, _score) in enumerate(vocab_scores)}
+            tokenizer = Tokenizer(
+                BPE(bpe_vocab, merges, unk_token=proto.trainer_spec.unk_piece, fuse_unk=True, byte_fallback=True)
+            )
+            tokenizer.add_special_tokens(
+                [ added_token for index, added_token in added_tokens.items()]
+            )
+        else:
+            raise Exception(
+                "You're trying to run a `Unigram` model but you're file was trained with a different algorithm"
+            )
+
+        return tokenizer
+
+    def normalizer(self, proto):
+        normalizers_list = []
+        if proto.normalizer_spec.add_dummy_prefix:
+            normalizers_list.append(normalizers.Prepend(prepend="▁"))
+        normalizers_list.append(normalizers.Replace(pattern=" ", content="▁"))
+        return normalizers.Sequence(normalizers_list)
+
+    def pre_tokenizer(self, replacement, add_prefix_space):
+        return None
+
+SLOW_TO_FAST_CONVERTERS["InternLM2Tokenizer"] = InternLM2Converter
+
+
+# Modified from transformers.model.llama.tokenization_llama_fast.LlamaTokenizerFast -> InternLM2TokenizerFast
+class InternLM2TokenizerFast(PreTrainedTokenizerFast):
+    vocab_files_names = VOCAB_FILES_NAMES
+    slow_tokenizer_class = InternLM2Tokenizer
+    padding_side = "left"
+    model_input_names = ["input_ids", "attention_mask"]
+    _auto_class = "AutoTokenizer"
+
+    def __init__(
+        self,
+        vocab_file,
+        unk_token="<unk>",
+        bos_token="<s>",
+        eos_token="</s>",
+        pad_token="</s>",
+        sp_model_kwargs: Optional[Dict[str, Any]] = None,
+        add_bos_token=True,
+        add_eos_token=False,
+        decode_with_prefix_space=False,
+        clean_up_tokenization_spaces=False,
+        **kwargs,
+    ):
+        super().__init__(
+            vocab_file=vocab_file,
+            unk_token=unk_token,
+            bos_token=bos_token,
+            eos_token=eos_token,
+            pad_token=pad_token,
+            sp_model_kwargs=sp_model_kwargs,
+            add_bos_token=add_bos_token,
+            add_eos_token=add_eos_token,
+            decode_with_prefix_space=decode_with_prefix_space,
+            clean_up_tokenization_spaces=clean_up_tokenization_spaces,
+            **kwargs,
+        )
+        self._add_bos_token = add_bos_token
+        self._add_eos_token = add_eos_token
+        self.update_post_processor()
+        self.vocab_file = vocab_file
+
+    @property
+    def can_save_slow_tokenizer(self) -> bool:
+        return os.path.isfile(self.vocab_file) if self.vocab_file else False
+
+    def update_post_processor(self):
+        """
+        Updates the underlying post processor with the current `bos_token` and `eos_token`.
+        """
+        bos = self.bos_token
+        bos_token_id = self.bos_token_id
+        if bos is None and self.add_bos_token:
+            raise ValueError("add_bos_token = True but bos_token = None")
+
+        eos = self.eos_token
+        eos_token_id = self.eos_token_id
+        if eos is None and self.add_eos_token:
+            raise ValueError("add_eos_token = True but eos_token = None")
+
+        single = f"{(bos+':0 ') if self.add_bos_token else ''}$A:0{(' '+eos+':0') if self.add_eos_token else ''}"
+        pair = f"{single}{(' '+bos+':1') if self.add_bos_token else ''} $B:1{(' '+eos+':1') if self.add_eos_token else ''}"
+
+        special_tokens = []
+        if self.add_bos_token:
+            special_tokens.append((bos, bos_token_id))
+        if self.add_eos_token:
+            special_tokens.append((eos, eos_token_id))
+        self._tokenizer.post_processor = processors.TemplateProcessing(
+            single=single, pair=pair, special_tokens=special_tokens
+        )
+
+    @property
+    def add_eos_token(self):
+        return self._add_eos_token
+
+    @property
+    def add_bos_token(self):
+        return self._add_bos_token
+
+    @add_eos_token.setter
+    def add_eos_token(self, value):
+        self._add_eos_token = value
+        self.update_post_processor()
+
+    @add_bos_token.setter
+    def add_bos_token(self, value):
+        self._add_bos_token = value
+        self.update_post_processor()
+
+    def save_vocabulary(self, save_directory: str, filename_prefix: Optional[str] = None) -> Tuple[str]:
+        if not self.can_save_slow_tokenizer:
+            raise ValueError(
+                "Your fast tokenizer does not have the necessary information to save the vocabulary for a slow "
+                "tokenizer."
+            )
+
+        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):
+            copyfile(self.vocab_file, out_vocab_file)
+
+        return (out_vocab_file,)

llama-factory/saves/internlm2_5_7b/lora/sft_bf16_p2_full/tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f868398fc4e05ee1e8aeba95ddf18ddcc45b8bce55d5093bead5bbf80429b48b
+size 1477754

llama-factory/saves/internlm2_5_7b/lora/sft_bf16_p2_full/tokenizer_config.json

@@ -0,0 +1,104 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "92538": {
+      "content": "<|plugin|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "92539": {
+      "content": "<|interpreter|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "92540": {
+      "content": "<|action_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "92541": {
+      "content": "<|action_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "92542": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "92543": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|action_start|>",
+    "<|action_end|>",
+    "<|interpreter|>",
+    "<|plugin|>"
+  ],
+  "auto_map": {
+    "AutoTokenizer": [
+      "tokenization_internlm2.InternLM2Tokenizer",
+      "tokenization_internlm2_fast.InternLM2TokenizerFast"
+    ]
+  },
+  "bos_token": "<s>",
+  "chat_template": "{{ '<s>' }}{% if messages[0]['role'] == 'system' %}{% set system_message = messages[0]['content'] %}{% endif %}{% if system_message is defined %}{{ '<|im_start|>system\n' + system_message + '<|im_end|>\n' }}{% endif %}{% for message in messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ '<|im_start|>user\n' + content + '<|im_end|>\n<|im_start|>assistant\n' }}{% elif message['role'] == 'assistant' %}{{ content + '<|im_end|>\n' }}{% endif %}{% endfor %}",
+  "clean_up_tokenization_spaces": false,
+  "decode_with_prefix_space": false,
+  "eos_token": "</s>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "</s>",
+  "padding_side": "right",
+  "sp_model_kwargs": null,
+  "split_special_tokens": false,
+  "tokenizer_class": "InternLM2Tokenizer",
+  "unk_token": "<unk>"
+}

llama-factory/saves/internlm2_5_7b/lora/sft_bf16_p2_full/train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 6.0,
+    "total_flos": 2.9041541335076045e+18,
+    "train_loss": 0.28717788867652416,
+    "train_runtime": 47077.7992,
+    "train_samples_per_second": 2.868,
+    "train_steps_per_second": 0.011
+}

llama-factory/saves/internlm2_5_7b/lora/sft_bf16_p2_full/trainer_log.jsonl

@@ -19,3 +19,21 @@
 {"current_steps": 352, "total_steps": 528, "eval_loss": 0.2945823669433594, "epoch": 4.0, "percentage": 66.67, "elapsed_time": "8:41:50", "remaining_time": "4:20:55", "throughput": "0.00", "total_tokens": 0}
 {"current_steps": 360, "total_steps": 528, "loss": 0.0856, "learning_rate": 2.7817622282960815e-05, "epoch": 4.090909090909091, "percentage": 68.18, "elapsed_time": "8:53:17", "remaining_time": "4:08:52", "throughput": "0.00", "total_tokens": 0}
 {"current_steps": 370, "total_steps": 528, "loss": 0.0661, "learning_rate": 2.490459770759398e-05, "epoch": 4.204545454545454, "percentage": 70.08, "elapsed_time": "9:07:34", "remaining_time": "3:53:49", "throughput": "0.00", "total_tokens": 0}
+{"current_steps": 380, "total_steps": 528, "loss": 0.061, "learning_rate": 2.2101308941239203e-05, "epoch": 4.318181818181818, "percentage": 71.97, "elapsed_time": "9:21:58", "remaining_time": "3:38:52", "throughput": "0.00", "total_tokens": 0}
+{"current_steps": 390, "total_steps": 528, "loss": 0.0744, "learning_rate": 1.942001405240979e-05, "epoch": 4.431818181818182, "percentage": 73.86, "elapsed_time": "9:36:17", "remaining_time": "3:23:55", "throughput": "0.00", "total_tokens": 0}
+{"current_steps": 400, "total_steps": 528, "loss": 0.0736, "learning_rate": 1.6872437661432517e-05, "epoch": 4.545454545454545, "percentage": 75.76, "elapsed_time": "9:50:45", "remaining_time": "3:09:02", "throughput": "0.00", "total_tokens": 0}
+{"current_steps": 410, "total_steps": 528, "loss": 0.0779, "learning_rate": 1.4469719671666043e-05, "epoch": 4.659090909090909, "percentage": 77.65, "elapsed_time": "10:05:19", "remaining_time": "2:54:12", "throughput": "0.00", "total_tokens": 0}
+{"current_steps": 420, "total_steps": 528, "loss": 0.075, "learning_rate": 1.2222366557537911e-05, "epoch": 4.7727272727272725, "percentage": 79.55, "elapsed_time": "10:19:52", "remaining_time": "2:39:23", "throughput": "0.00", "total_tokens": 0}
+{"current_steps": 430, "total_steps": 528, "loss": 0.0752, "learning_rate": 1.0140205422405214e-05, "epoch": 4.886363636363637, "percentage": 81.44, "elapsed_time": "10:34:26", "remaining_time": "2:24:35", "throughput": "0.00", "total_tokens": 0}
+{"current_steps": 440, "total_steps": 528, "loss": 0.0684, "learning_rate": 8.232341027131885e-06, "epoch": 5.0, "percentage": 83.33, "elapsed_time": "10:48:47", "remaining_time": "2:09:45", "throughput": "0.00", "total_tokens": 0}
+{"current_steps": 440, "total_steps": 528, "eval_loss": 0.369967520236969, "epoch": 5.0, "percentage": 83.33, "elapsed_time": "10:52:39", "remaining_time": "2:10:31", "throughput": "0.00", "total_tokens": 0}
+{"current_steps": 450, "total_steps": 528, "loss": 0.0531, "learning_rate": 6.5071159772861436e-06, "epoch": 5.113636363636363, "percentage": 85.23, "elapsed_time": "11:07:14", "remaining_time": "1:55:39", "throughput": "0.00", "total_tokens": 0}
+{"current_steps": 460, "total_steps": 528, "loss": 0.0437, "learning_rate": 4.972074243048897e-06, "epoch": 5.2272727272727275, "percentage": 87.12, "elapsed_time": "11:21:48", "remaining_time": "1:40:47", "throughput": "0.00", "total_tokens": 0}
+{"current_steps": 470, "total_steps": 528, "loss": 0.0463, "learning_rate": 3.6339281713517303e-06, "epoch": 5.340909090909091, "percentage": 89.02, "elapsed_time": "11:36:22", "remaining_time": "1:25:56", "throughput": "0.00", "total_tokens": 0}
+{"current_steps": 480, "total_steps": 528, "loss": 0.0485, "learning_rate": 2.4985291344915674e-06, "epoch": 5.454545454545454, "percentage": 90.91, "elapsed_time": "11:50:53", "remaining_time": "1:11:05", "throughput": "0.00", "total_tokens": 0}
+{"current_steps": 490, "total_steps": 528, "loss": 0.0495, "learning_rate": 1.5708419435684462e-06, "epoch": 5.568181818181818, "percentage": 92.8, "elapsed_time": "12:05:26", "remaining_time": "0:56:15", "throughput": "0.00", "total_tokens": 0}
+{"current_steps": 500, "total_steps": 528, "loss": 0.0484, "learning_rate": 8.549231386298151e-07, "epoch": 5.681818181818182, "percentage": 94.7, "elapsed_time": "12:20:01", "remaining_time": "0:41:26", "throughput": "0.00", "total_tokens": 0}
+{"current_steps": 510, "total_steps": 528, "loss": 0.0384, "learning_rate": 3.5390325045304706e-07, "epoch": 5.795454545454545, "percentage": 96.59, "elapsed_time": "12:34:32", "remaining_time": "0:26:37", "throughput": "0.00", "total_tokens": 0}
+{"current_steps": 520, "total_steps": 528, "loss": 0.0486, "learning_rate": 6.997311153086883e-08, "epoch": 5.909090909090909, "percentage": 98.48, "elapsed_time": "12:49:04", "remaining_time": "0:11:49", "throughput": "0.00", "total_tokens": 0}
+{"current_steps": 528, "total_steps": 528, "eval_loss": 0.4368518590927124, "epoch": 6.0, "percentage": 100.0, "elapsed_time": "13:04:27", "remaining_time": "0:00:00", "throughput": "0.00", "total_tokens": 0}
+{"current_steps": 528, "total_steps": 528, "epoch": 6.0, "percentage": 100.0, "elapsed_time": "13:04:28", "remaining_time": "0:00:00", "throughput": "0.00", "total_tokens": 0}

llama-factory/saves/internlm2_5_7b/lora/sft_bf16_p2_full/trainer_state.json

@@ -0,0 +1,460 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 6.0,
+  "eval_steps": 88,
+  "global_step": 528,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.11363636363636363,
+      "grad_norm": 3.514838457107544,
+      "learning_rate": 1.8867924528301888e-05,
+      "loss": 6.943,
+      "step": 10
+    },
+    {
+      "epoch": 0.22727272727272727,
+      "grad_norm": 1.0595427751541138,
+      "learning_rate": 3.7735849056603776e-05,
+      "loss": 0.446,
+      "step": 20
+    },
+    {
+      "epoch": 0.3409090909090909,
+      "grad_norm": 0.6256385445594788,
+      "learning_rate": 5.660377358490566e-05,
+      "loss": 0.3515,
+      "step": 30
+    },
+    {
+      "epoch": 0.45454545454545453,
+      "grad_norm": 0.633573055267334,
+      "learning_rate": 7.547169811320755e-05,
+      "loss": 0.288,
+      "step": 40
+    },
+    {
+      "epoch": 0.5681818181818182,
+      "grad_norm": 0.4915701746940613,
+      "learning_rate": 9.433962264150944e-05,
+      "loss": 0.2819,
+      "step": 50
+    },
+    {
+      "epoch": 0.6818181818181818,
+      "grad_norm": 0.40083640813827515,
+      "learning_rate": 9.994642390694308e-05,
+      "loss": 0.2765,
+      "step": 60
+    },
+    {
+      "epoch": 0.7954545454545454,
+      "grad_norm": 0.7176418304443359,
+      "learning_rate": 9.968428675226714e-05,
+      "loss": 0.2754,
+      "step": 70
+    },
+    {
+      "epoch": 0.9090909090909091,
+      "grad_norm": 0.6853049397468567,
+      "learning_rate": 9.92048928531717e-05,
+      "loss": 0.277,
+      "step": 80
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.9030666666666668,
+      "eval_loss": 0.26003387570381165,
+      "eval_runtime": 231.9406,
+      "eval_samples_per_second": 10.779,
+      "eval_steps_per_second": 10.779,
+      "step": 88
+    },
+    {
+      "epoch": 1.0227272727272727,
+      "grad_norm": 0.4268323481082916,
+      "learning_rate": 9.851033847720166e-05,
+      "loss": 0.2619,
+      "step": 90
+    },
+    {
+      "epoch": 1.1363636363636362,
+      "grad_norm": 0.9503114819526672,
+      "learning_rate": 9.760366073392246e-05,
+      "loss": 0.2385,
+      "step": 100
+    },
+    {
+      "epoch": 1.25,
+      "grad_norm": 0.3606574237346649,
+      "learning_rate": 9.648882429441257e-05,
+      "loss": 0.2341,
+      "step": 110
+    },
+    {
+      "epoch": 1.3636363636363638,
+      "grad_norm": 0.7226484417915344,
+      "learning_rate": 9.517070405476575e-05,
+      "loss": 0.2447,
+      "step": 120
+    },
+    {
+      "epoch": 1.4772727272727273,
+      "grad_norm": 0.8543397188186646,
+      "learning_rate": 9.365506381941066e-05,
+      "loss": 0.2441,
+      "step": 130
+    },
+    {
+      "epoch": 1.5909090909090908,
+      "grad_norm": 0.800394594669342,
+      "learning_rate": 9.194853109746074e-05,
+      "loss": 0.2379,
+      "step": 140
+    },
+    {
+      "epoch": 1.7045454545454546,
+      "grad_norm": 0.5756838321685791,
+      "learning_rate": 9.005856812230304e-05,
+      "loss": 0.2434,
+      "step": 150
+    },
+    {
+      "epoch": 1.8181818181818183,
+      "grad_norm": 1.0771032571792603,
+      "learning_rate": 8.799343922115044e-05,
+      "loss": 0.2352,
+      "step": 160
+    },
+    {
+      "epoch": 1.9318181818181817,
+      "grad_norm": 0.4805872440338135,
+      "learning_rate": 8.576217467724128e-05,
+      "loss": 0.2427,
+      "step": 170
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.9006,
+      "eval_loss": 0.25738978385925293,
+      "eval_runtime": 231.2131,
+      "eval_samples_per_second": 10.813,
+      "eval_steps_per_second": 10.813,
+      "step": 176
+    },
+    {
+      "epoch": 2.0454545454545454,
+      "grad_norm": 0.5219587683677673,
+      "learning_rate": 8.337453124270863e-05,
+      "loss": 0.22,
+      "step": 180
+    },
+    {
+      "epoch": 2.159090909090909,
+      "grad_norm": 0.6363154053688049,
+      "learning_rate": 8.084094947478556e-05,
+      "loss": 0.1787,
+      "step": 190
+    },
+    {
+      "epoch": 2.2727272727272725,
+      "grad_norm": 0.6807820796966553,
+      "learning_rate": 7.817250808190483e-05,
+      "loss": 0.1647,
+      "step": 200
+    },
+    {
+      "epoch": 2.3863636363636362,
+      "grad_norm": 0.5443515777587891,
+      "learning_rate": 7.538087547932585e-05,
+      "loss": 0.1828,
+      "step": 210
+    },
+    {
+      "epoch": 2.5,
+      "grad_norm": 0.4641902446746826,
+      "learning_rate": 7.247825876612353e-05,
+      "loss": 0.1782,
+      "step": 220
+    },
+    {
+      "epoch": 2.6136363636363638,
+      "grad_norm": 0.5865933299064636,
+      "learning_rate": 6.947735034665002e-05,
+      "loss": 0.1942,
+      "step": 230
+    },
+    {
+      "epoch": 2.7272727272727275,
+      "grad_norm": 0.5332173705101013,
+      "learning_rate": 6.639127242987988e-05,
+      "loss": 0.1852,
+      "step": 240
+    },
+    {
+      "epoch": 2.840909090909091,
+      "grad_norm": 0.5550218820571899,
+      "learning_rate": 6.323351964932908e-05,
+      "loss": 0.1936,
+      "step": 250
+    },
+    {
+      "epoch": 2.9545454545454546,
+      "grad_norm": 0.6850063800811768,
+      "learning_rate": 6.001790005445607e-05,
+      "loss": 0.1813,
+      "step": 260
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.9027,
+      "eval_loss": 0.2705931067466736,
+      "eval_runtime": 231.0113,
+      "eval_samples_per_second": 10.822,
+      "eval_steps_per_second": 10.822,
+      "step": 264
+    },
+    {
+      "epoch": 3.0681818181818183,
+      "grad_norm": 0.42733630537986755,
+      "learning_rate": 5.675847473157485e-05,
+      "loss": 0.14,
+      "step": 270
+    },
+    {
+      "epoch": 3.1818181818181817,
+      "grad_norm": 0.5972977876663208,
+      "learning_rate": 5.3469496318302204e-05,
+      "loss": 0.1197,
+      "step": 280
+    },
+    {
+      "epoch": 3.2954545454545454,
+      "grad_norm": 0.4995785653591156,
+      "learning_rate": 5.016534668039976e-05,
+      "loss": 0.1198,
+      "step": 290
+    },
+    {
+      "epoch": 3.409090909090909,
+      "grad_norm": 0.5500032305717468,
+      "learning_rate": 4.6860474023534335e-05,
+      "loss": 0.1131,
+      "step": 300
+    },
+    {
+      "epoch": 3.5227272727272725,
+      "grad_norm": 0.4452584683895111,
+      "learning_rate": 4.3569329714950704e-05,
+      "loss": 0.1185,
+      "step": 310
+    },
+    {
+      "epoch": 3.6363636363636362,
+      "grad_norm": 0.4754205346107483,
+      "learning_rate": 4.0306305091319595e-05,
+      "loss": 0.1197,
+      "step": 320
+    },
+    {
+      "epoch": 3.75,
+      "grad_norm": 0.6347799301147461,
+      "learning_rate": 3.7085668529084184e-05,
+      "loss": 0.122,
+      "step": 330
+    },
+    {
+      "epoch": 3.8636363636363638,
+      "grad_norm": 0.48911160230636597,
+      "learning_rate": 3.392150305248024e-05,
+      "loss": 0.1163,
+      "step": 340
+    },
+    {
+      "epoch": 3.9772727272727275,
+      "grad_norm": 0.6460514068603516,
+      "learning_rate": 3.082764475205442e-05,
+      "loss": 0.1263,
+      "step": 350
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.8993666666666668,
+      "eval_loss": 0.2945823669433594,
+      "eval_runtime": 233.0443,
+      "eval_samples_per_second": 10.728,
+      "eval_steps_per_second": 10.728,
+      "step": 352
+    },
+    {
+      "epoch": 4.090909090909091,
+      "grad_norm": 0.37293320894241333,
+      "learning_rate": 2.7817622282960815e-05,
+      "loss": 0.0856,
+      "step": 360
+    },
+    {
+      "epoch": 4.204545454545454,
+      "grad_norm": 0.5676562190055847,
+      "learning_rate": 2.490459770759398e-05,
+      "loss": 0.0661,
+      "step": 370
+    },
+    {
+      "epoch": 4.318181818181818,
+      "grad_norm": 0.5680781006813049,
+      "learning_rate": 2.2101308941239203e-05,
+      "loss": 0.061,
+      "step": 380
+    },
+    {
+      "epoch": 4.431818181818182,
+      "grad_norm": 0.690169095993042,
+      "learning_rate": 1.942001405240979e-05,
+      "loss": 0.0744,
+      "step": 390
+    },
+    {
+      "epoch": 4.545454545454545,
+      "grad_norm": 0.5858839750289917,
+      "learning_rate": 1.6872437661432517e-05,
+      "loss": 0.0736,
+      "step": 400
+    },
+    {
+      "epoch": 4.659090909090909,
+      "grad_norm": 0.6473811268806458,
+      "learning_rate": 1.4469719671666043e-05,
+      "loss": 0.0779,
+      "step": 410
+    },
+    {
+      "epoch": 4.7727272727272725,
+      "grad_norm": 0.3694300055503845,
+      "learning_rate": 1.2222366557537911e-05,
+      "loss": 0.075,
+      "step": 420
+    },
+    {
+      "epoch": 4.886363636363637,
+      "grad_norm": 0.5935441851615906,
+      "learning_rate": 1.0140205422405214e-05,
+      "loss": 0.0752,
+      "step": 430
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 0.7272607684135437,
+      "learning_rate": 8.232341027131885e-06,
+      "loss": 0.0684,
+      "step": 440
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.8996666666666668,
+      "eval_loss": 0.369967520236969,
+      "eval_runtime": 232.4252,
+      "eval_samples_per_second": 10.756,
+      "eval_steps_per_second": 10.756,
+      "step": 440
+    },
+    {
+      "epoch": 5.113636363636363,
+      "grad_norm": 0.3855780363082886,
+      "learning_rate": 6.5071159772861436e-06,
+      "loss": 0.0531,
+      "step": 450
+    },
+    {
+      "epoch": 5.2272727272727275,
+      "grad_norm": 0.3242223560810089,
+      "learning_rate": 4.972074243048897e-06,
+      "loss": 0.0437,
+      "step": 460
+    },
+    {
+      "epoch": 5.340909090909091,
+      "grad_norm": 0.36955130100250244,
+      "learning_rate": 3.6339281713517303e-06,
+      "loss": 0.0463,
+      "step": 470
+    },
+    {
+      "epoch": 5.454545454545454,
+      "grad_norm": 0.3851165473461151,
+      "learning_rate": 2.4985291344915674e-06,
+      "loss": 0.0485,
+      "step": 480
+    },
+    {
+      "epoch": 5.568181818181818,
+      "grad_norm": 0.30520951747894287,
+      "learning_rate": 1.5708419435684462e-06,
+      "loss": 0.0495,
+      "step": 490
+    },
+    {
+      "epoch": 5.681818181818182,
+      "grad_norm": 0.8094011545181274,
+      "learning_rate": 8.549231386298151e-07,
+      "loss": 0.0484,
+      "step": 500
+    },
+    {
+      "epoch": 5.795454545454545,
+      "grad_norm": 0.21888971328735352,
+      "learning_rate": 3.5390325045304706e-07,
+      "loss": 0.0384,
+      "step": 510
+    },
+    {
+      "epoch": 5.909090909090909,
+      "grad_norm": 0.4017506539821625,
+      "learning_rate": 6.997311153086883e-08,
+      "loss": 0.0486,
+      "step": 520
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.8984,
+      "eval_loss": 0.4368518590927124,
+      "eval_runtime": 232.9443,
+      "eval_samples_per_second": 10.732,
+      "eval_steps_per_second": 10.732,
+      "step": 528
+    },
+    {
+      "epoch": 6.0,
+      "step": 528,
+      "total_flos": 2.9041541335076045e+18,
+      "train_loss": 0.28717788867652416,
+      "train_runtime": 47077.7992,
+      "train_samples_per_second": 2.868,
+      "train_steps_per_second": 0.011
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 528,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 6,
+  "save_steps": 88,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 2.9041541335076045e+18,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}

llama-factory/saves/internlm2_5_7b/lora/sft_bf16_p2_full/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9d5c2bfb768f1720511bef6e983b38242156f823ff83caba648f7520ad9d664c
+size 5304