初始化项目，由ModelHub XC社区提供模型

Model: lldois/v24_v15_cot_light_repair_lr25e6_ep035 Source: Original Platform
2026-07-18 07:37:10 +08:00
commit 3e0527763f
17 changed files with 2734 additions and 0 deletions
--- a/.gitattributes
+++ b/.gitattributes
@@ -0,0 +1,36 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text
--- a/README.md
+++ b/README.md
@@ -0,0 +1,58 @@
+---
+library_name: transformers
+license: other
+base_model: lldois/v15_user_logic_json_lr15e6
+tags:
+- llama-factory
+- full
+- generated_from_trainer
+model-index:
+- name: v24_v15_cot_light_repair_lr25e6_ep035
+  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. -->
+
+# v24_v15_cot_light_repair_lr25e6_ep035
+
+This model is a fine-tuned version of [lldois/v15_user_logic_json_lr15e6](https://huggingface.co/lldois/v15_user_logic_json_lr15e6) on the exp_v15_cot_light_repair dataset.
+
+## 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: 2.5e-06
+- train_batch_size: 1
+- eval_batch_size: 8
+- seed: 202607243
+- gradient_accumulation_steps: 4
+- total_train_batch_size: 4
+- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_steps: 0.02
+- num_epochs: 0.35
+
+### Training results
+
+
+
+### Framework versions
+
+- Transformers 5.6.0
+- Pytorch 2.7.1+cu126
+- Datasets 4.0.0
+- Tokenizers 0.22.2
--- a/all_results.json
+++ b/all_results.json
@@ -0,0 +1,8 @@
+{
+    "epoch": 0.351575456053068,
+    "total_flos": 5.062818542541619e+16,
+    "train_loss": 1.7334247456406646,
+    "train_runtime": 876.5295,
+    "train_samples_per_second": 0.482,
+    "train_steps_per_second": 0.121
+}
--- a/chat_template.jinja
+++ b/chat_template.jinja
@@ -0,0 +1,89 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0].role == 'system' %}
+        {{- messages[0].content + '\n\n' }}
+    {%- endif %}
+    {{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
+{%- else %}
+    {%- if messages[0].role == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
+{%- for message in messages[::-1] %}
+    {%- set index = (messages|length - 1) - loop.index0 %}
+    {%- if ns.multi_step_tool and message.role == "user" and message.content is string and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
+        {%- set ns.multi_step_tool = false %}
+        {%- set ns.last_query_index = index %}
+    {%- endif %}
+{%- endfor %}
+{%- for message in messages %}
+    {%- if message.content is string %}
+        {%- set content = message.content %}
+    {%- else %}
+        {%- set content = '' %}
+    {%- endif %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
+        {{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is string %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- else %}
+            {%- if '</think>' in content %}
+                {%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+                {%- set content = content.split('</think>')[-1].lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {%- if loop.index0 > ns.last_query_index %}
+            {%- if loop.last or (not loop.last and reasoning_content) %}
+                {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
+            {%- else %}
+                {{- '<|im_start|>' + message.role + '\n' + content }}
+            {%- endif %}
+        {%- else %}
+            {{- '<|im_start|>' + message.role + '\n' + content }}
+        {%- endif %}
+        {%- if message.tool_calls %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if (loop.first and content) or (not loop.first) %}
+                    {{- '\n' }}
+                {%- endif %}
+                {%- if tool_call.function %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {{- '<tool_call>\n{"name": "' }}
+                {{- tool_call.name }}
+                {{- '", "arguments": ' }}
+                {%- if tool_call.arguments is string %}
+                    {{- tool_call.arguments }}
+                {%- else %}
+                    {{- tool_call.arguments | tojson }}
+                {%- endif %}
+                {{- '}\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+    {%- if enable_thinking is defined and enable_thinking is false %}
+        {{- '<think>\n\n</think>\n\n' }}
+    {%- endif %}
+{%- endif %}
--- a/config.json
+++ b/config.json
@@ -0,0 +1,63 @@
+{
+  "architectures": [
+    "Qwen3ForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": null,
+  "dtype": "bfloat16",
+  "eos_token_id": 151645,
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "hidden_size": 1024,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_types": [
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention"
+  ],
+  "max_position_embeddings": 40960,
+  "max_window_layers": 28,
+  "model_type": "qwen3",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 28,
+  "num_key_value_heads": 8,
+  "pad_token_id": 151643,
+  "rms_norm_eps": 1e-06,
+  "rope_parameters": {
+    "rope_theta": 1000000,
+    "rope_type": "default"
+  },
+  "sliding_window": null,
+  "tie_word_embeddings": false,
+  "transformers_version": "5.6.0",
+  "use_cache": false,
+  "use_sliding_window": false,
+  "vocab_size": 176253
+}
--- a/experiment_recipe.json
+++ b/experiment_recipe.json
@@ -0,0 +1,59 @@
+{
+  "created_at": "2026-07-10 01:00:49",
+  "run": {
+    "name": "v24_v15_cot_light_repair_lr25e6_ep035",
+    "dataset": "v15_cot_light_repair",
+    "model_path": "/home/ll/llm4rec/experiments/outputs/v15_user_logic_json_lr15e6",
+    "lr": "2.5e-6",
+    "epochs": 0.35,
+    "warmup": 0.02,
+    "scheduler": "cosine",
+    "seed": 202607243,
+    "note": "v15 continuation. Small CoT-native repair with strong item replay, strict user JSON, and limited rec fast/short-CoT supervision. Goal: improve v15/v19 while avoiding v19 world loss and without using v7 as a CoT base.",
+    "config": "/home/ll/llm4rec/experiments/configs/v24_v15_cot_light_repair_lr25e6_ep035.yaml"
+  },
+  "dataset_manifest": {
+    "name": "v15_cot_light_repair",
+    "path": "/home/ll/llm4rec/experiments/data/v15_cot_light_repair.jsonl",
+    "records": 27334,
+    "groups": {
+      "rec": 11200,
+      "user": 4940,
+      "item": 11194
+    },
+    "variants": {
+      "rec_short_think_final": 4800,
+      "user_strict_array": 2375,
+      "rec_no_think_direct_final": 6400,
+      "item_no_think_direct_final": 2805,
+      "user_strict_logic": 1963,
+      "item_compact_cot": 5597,
+      "user_extra_no_think_logic": 602,
+      "item_short_think": 2792
+    },
+    "sha256": "2842c6490bff3273d3b15c4443bd1ea56ef7806c9d7f1800110acca5a521de19",
+    "seed": 202607113
+  },
+  "dataset_recipe": "v15 continuation: 6400 rec /no_think direct-final + 4800 short rec /think CoT + strict dual user aug + half strict user replay + item route aug + compact item CoT aug.",
+  "cot_policy": "Preserve /think reasoning supervision and do not use v7_final_only as a CoT training base. For /no_think prompts, train pure final answers without generated <think> tags. This is route-specific behavior, not global CoT removal.",
+  "raw_counts": {
+    "rec": 19204,
+    "item": 10384,
+    "user": 2892
+  },
+  "eval_observations": {
+    "v07": "best local score so far: total=0.8978, eval_time≈47.3min; fast final outputs likely help.",
+    "v15": "best CoT-preserving score so far: total=0.8778, eval_time≈70.1min; logs show repeated tokens, JSON shell errors, prompt leakage, and verbose /no_think outputs.",
+    "v19": "best CoT-native continuation so far: total=0.8855, eval_time≈48.1min; user1 and rec4 improved but world dropped.",
+    "v20": "v7 final-only continuation with CoT restore failed as a CoT route: total=0.8527, item fell to 0.1840; do not use v7 as future CoT base.",
+    "v16": "CoT pattern rewrite failed: total=0.7912; logs show malformed user JSON and fragmented recommendation reasoning.",
+    "v18": "low-LR mixed replay from v12 failed: total=0.8340; item/world dropped and rec outputs mixed text/itemic/think tags."
+  },
+  "script": "/home/ll/llm4rec/experiments/run_experiments.py",
+  "script_sha256": "a03ddf5a479db437247aed4b8000b25cb5021988029220bd818133123951aa00",
+  "deadline": "2026-07-10 09:00:00 +0800",
+  "reproduce": {
+    "prepare_command": "EXPERIMENT_PREPARE_ONLY=1 python3 experiments/run_experiments.py",
+    "train_command": "CUDA_VISIBLE_DEVICES=<gpu> bash -lc 'source /home/ll/llm4rec/demo/LLaMA-Factory/.venv/bin/activate && llamafactory-cli train /home/ll/llm4rec/experiments/configs/v24_v15_cot_light_repair_lr25e6_ep035.yaml'"
+  }
+}
--- a/generation_config.json
+++ b/generation_config.json
@@ -0,0 +1,12 @@
+{
+  "do_sample": true,
+  "eos_token_id": [
+    151645,
+    151643
+  ],
+  "pad_token_id": 151643,
+  "temperature": 0.6,
+  "top_k": 20,
+  "top_p": 0.95,
+  "transformers_version": "5.6.0"
+}
--- a/model.safetensors
+++ b/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:35b4f52c029a4dd2d53c4b933ef2847bf407b97b64326d71f0209c1ca1b2422a
+size 1602902832
--- a/runs/Jul10_01-01-16_ubuntu-System-Product-Name/events.out.tfevents.1783616476.ubuntu-System-Product-Name.2039687.0
+++ b/runs/Jul10_01-01-16_ubuntu-System-Product-Name/events.out.tfevents.1783616476.ubuntu-System-Product-Name.2039687.0
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5eb82ab745d62f612d2f86faa1a38f8559ccd47757be75d9b39b583ba4e7b160
+size 10655
--- a/tokenizer.json
+++ b/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cd4d15f596979aecbc11ba12668cb21c9fb8452ce1235cd9d7878cc950170421
+size 16016554
--- a/tokenizer_config.json
+++ b/tokenizer_config.json
@@ -0,0 +1,16 @@
+{
+  "add_prefix_space": false,
+  "backend": "tokenizers",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "is_local": true,
+  "local_files_only": false,
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}
--- a/train.log
+++ b/train.log
--- a/train_results.json
+++ b/train_results.json
@@ -0,0 +1,8 @@
+{
+    "epoch": 0.351575456053068,
+    "total_flos": 5.062818542541619e+16,
+    "train_loss": 1.7334247456406646,
+    "train_runtime": 876.5295,
+    "train_samples_per_second": 0.482,
+    "train_steps_per_second": 0.121
+}
--- a/trainer_log.jsonl
+++ b/trainer_log.jsonl
@@ -0,0 +1,22 @@
+{"current_steps": 5, "total_steps": 106, "loss": 1.9901918411254882, "lr": 2.4994186045291896e-06, "epoch": 0.01658374792703151, "percentage": 4.72, "elapsed_time": "0:00:42", "remaining_time": "0:14:09"}
+{"current_steps": 10, "total_steps": 106, "loss": 2.094209098815918, "lr": 2.4791264942053156e-06, "epoch": 0.03316749585406302, "percentage": 9.43, "elapsed_time": "0:01:23", "remaining_time": "0:13:17"}
+{"current_steps": 15, "total_steps": 106, "loss": 1.80328426361084, "lr": 2.4303031863746234e-06, "epoch": 0.04975124378109453, "percentage": 14.15, "elapsed_time": "0:02:03", "remaining_time": "0:12:31"}
+{"current_steps": 20, "total_steps": 106, "loss": 2.002113914489746, "lr": 2.354081995993076e-06, "epoch": 0.06633499170812604, "percentage": 18.87, "elapsed_time": "0:02:44", "remaining_time": "0:11:45"}
+{"current_steps": 25, "total_steps": 106, "loss": 1.559170913696289, "lr": 2.2522322135667634e-06, "epoch": 0.08291873963515754, "percentage": 23.58, "elapsed_time": "0:03:27", "remaining_time": "0:11:11"}
+{"current_steps": 30, "total_steps": 106, "loss": 1.7275423049926757, "lr": 2.127118035353867e-06, "epoch": 0.09950248756218906, "percentage": 28.3, "elapsed_time": "0:04:07", "remaining_time": "0:10:26"}
+{"current_steps": 35, "total_steps": 106, "loss": 1.4820417404174804, "lr": 1.9816436842690527e-06, "epoch": 0.11608623548922056, "percentage": 33.02, "elapsed_time": "0:04:50", "remaining_time": "0:09:48"}
+{"current_steps": 40, "total_steps": 106, "loss": 1.4906288146972657, "lr": 1.819185995375714e-06, "epoch": 0.13266998341625208, "percentage": 37.74, "elapsed_time": "0:05:33", "remaining_time": "0:09:10"}
+{"current_steps": 45, "total_steps": 106, "loss": 1.7329635620117188, "lr": 1.6435160308307313e-06, "epoch": 0.14925373134328357, "percentage": 42.45, "elapsed_time": "0:06:15", "remaining_time": "0:08:28"}
+{"current_steps": 50, "total_steps": 106, "loss": 1.8760557174682617, "lr": 1.458711543801153e-06, "epoch": 0.16583747927031509, "percentage": 47.17, "elapsed_time": "0:06:56", "remaining_time": "0:07:46"}
+{"current_steps": 55, "total_steps": 106, "loss": 1.546410369873047, "lr": 1.2690623232911575e-06, "epoch": 0.1824212271973466, "percentage": 51.89, "elapsed_time": "0:07:38", "remaining_time": "0:07:05"}
+{"current_steps": 60, "total_steps": 106, "loss": 1.8373964309692383, "lr": 1.0789706170701125e-06, "epoch": 0.19900497512437812, "percentage": 56.6, "elapsed_time": "0:08:19", "remaining_time": "0:06:23"}
+{"current_steps": 65, "total_steps": 106, "loss": 2.1412117004394533, "lr": 8.928489441425056e-07, "epoch": 0.2155887230514096, "percentage": 61.32, "elapsed_time": "0:08:58", "remaining_time": "0:05:39"}
+{"current_steps": 70, "total_steps": 106, "loss": 1.5714065551757812, "lr": 7.15017668796009e-07, "epoch": 0.23217247097844113, "percentage": 66.04, "elapsed_time": "0:09:40", "remaining_time": "0:04:58"}
+{"current_steps": 75, "total_steps": 106, "loss": 1.6369653701782227, "lr": 5.496047137983128e-07, "epoch": 0.24875621890547264, "percentage": 70.75, "elapsed_time": "0:10:22", "remaining_time": "0:04:17"}
+{"current_steps": 80, "total_steps": 106, "loss": 1.8116127014160157, "lr": 4.0044974065819974e-07, "epoch": 0.26533996683250416, "percentage": 75.47, "elapsed_time": "0:11:00", "remaining_time": "0:03:34"}
+{"current_steps": 85, "total_steps": 106, "loss": 1.5387956619262695, "lr": 2.7101502117425833e-07, "epoch": 0.28192371475953565, "percentage": 80.19, "elapsed_time": "0:11:41", "remaining_time": "0:02:53"}
+{"current_steps": 90, "total_steps": 106, "loss": 1.850395393371582, "lr": 1.6430506917256422e-07, "epoch": 0.29850746268656714, "percentage": 84.91, "elapsed_time": "0:12:21", "remaining_time": "0:02:11"}
+{"current_steps": 95, "total_steps": 106, "loss": 1.588356113433838, "lr": 8.279689798808068e-08, "epoch": 0.3150912106135987, "percentage": 89.62, "elapsed_time": "0:13:02", "remaining_time": "0:01:30"}
+{"current_steps": 100, "total_steps": 106, "loss": 1.513393497467041, "lr": 2.8382522593593807e-08, "epoch": 0.33167495854063017, "percentage": 94.34, "elapsed_time": "0:13:45", "remaining_time": "0:00:49"}
+{"current_steps": 105, "total_steps": 106, "loss": 1.58885498046875, "lr": 2.3250410501327037e-09, "epoch": 0.3482587064676617, "percentage": 99.06, "elapsed_time": "0:14:28", "remaining_time": "0:00:08"}
+{"current_steps": 106, "total_steps": 106, "epoch": 0.351575456053068, "percentage": 100.0, "elapsed_time": "0:14:36", "remaining_time": "0:00:00"}
--- a/trainer_state.json
+++ b/trainer_state.json
@@ -0,0 +1,190 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.351575456053068,
+  "eval_steps": 500,
+  "global_step": 106,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.01658374792703151,
+      "grad_norm": 26.625,
+      "learning_rate": 2.4994186045291896e-06,
+      "loss": 1.9901918411254882,
+      "step": 5
+    },
+    {
+      "epoch": 0.03316749585406302,
+      "grad_norm": 21.75,
+      "learning_rate": 2.4791264942053156e-06,
+      "loss": 2.094209098815918,
+      "step": 10
+    },
+    {
+      "epoch": 0.04975124378109453,
+      "grad_norm": 19.25,
+      "learning_rate": 2.4303031863746234e-06,
+      "loss": 1.80328426361084,
+      "step": 15
+    },
+    {
+      "epoch": 0.06633499170812604,
+      "grad_norm": 26.375,
+      "learning_rate": 2.354081995993076e-06,
+      "loss": 2.002113914489746,
+      "step": 20
+    },
+    {
+      "epoch": 0.08291873963515754,
+      "grad_norm": 20.375,
+      "learning_rate": 2.2522322135667634e-06,
+      "loss": 1.559170913696289,
+      "step": 25
+    },
+    {
+      "epoch": 0.09950248756218906,
+      "grad_norm": 12.6875,
+      "learning_rate": 2.127118035353867e-06,
+      "loss": 1.7275423049926757,
+      "step": 30
+    },
+    {
+      "epoch": 0.11608623548922056,
+      "grad_norm": 23.125,
+      "learning_rate": 1.9816436842690527e-06,
+      "loss": 1.4820417404174804,
+      "step": 35
+    },
+    {
+      "epoch": 0.13266998341625208,
+      "grad_norm": 10.8125,
+      "learning_rate": 1.819185995375714e-06,
+      "loss": 1.4906288146972657,
+      "step": 40
+    },
+    {
+      "epoch": 0.14925373134328357,
+      "grad_norm": 12.6875,
+      "learning_rate": 1.6435160308307313e-06,
+      "loss": 1.7329635620117188,
+      "step": 45
+    },
+    {
+      "epoch": 0.16583747927031509,
+      "grad_norm": 36.75,
+      "learning_rate": 1.458711543801153e-06,
+      "loss": 1.8760557174682617,
+      "step": 50
+    },
+    {
+      "epoch": 0.1824212271973466,
+      "grad_norm": 10.5625,
+      "learning_rate": 1.2690623232911575e-06,
+      "loss": 1.546410369873047,
+      "step": 55
+    },
+    {
+      "epoch": 0.19900497512437812,
+      "grad_norm": 46.25,
+      "learning_rate": 1.0789706170701125e-06,
+      "loss": 1.8373964309692383,
+      "step": 60
+    },
+    {
+      "epoch": 0.2155887230514096,
+      "grad_norm": 54.25,
+      "learning_rate": 8.928489441425056e-07,
+      "loss": 2.1412117004394533,
+      "step": 65
+    },
+    {
+      "epoch": 0.23217247097844113,
+      "grad_norm": 17.75,
+      "learning_rate": 7.15017668796009e-07,
+      "loss": 1.5714065551757812,
+      "step": 70
+    },
+    {
+      "epoch": 0.24875621890547264,
+      "grad_norm": 10.8125,
+      "learning_rate": 5.496047137983128e-07,
+      "loss": 1.6369653701782227,
+      "step": 75
+    },
+    {
+      "epoch": 0.26533996683250416,
+      "grad_norm": 17.625,
+      "learning_rate": 4.0044974065819974e-07,
+      "loss": 1.8116127014160157,
+      "step": 80
+    },
+    {
+      "epoch": 0.28192371475953565,
+      "grad_norm": 12.1875,
+      "learning_rate": 2.7101502117425833e-07,
+      "loss": 1.5387956619262695,
+      "step": 85
+    },
+    {
+      "epoch": 0.29850746268656714,
+      "grad_norm": 12.3125,
+      "learning_rate": 1.6430506917256422e-07,
+      "loss": 1.850395393371582,
+      "step": 90
+    },
+    {
+      "epoch": 0.3150912106135987,
+      "grad_norm": 7.78125,
+      "learning_rate": 8.279689798808068e-08,
+      "loss": 1.588356113433838,
+      "step": 95
+    },
+    {
+      "epoch": 0.33167495854063017,
+      "grad_norm": 7.0,
+      "learning_rate": 2.8382522593593807e-08,
+      "loss": 1.513393497467041,
+      "step": 100
+    },
+    {
+      "epoch": 0.3482587064676617,
+      "grad_norm": 56.75,
+      "learning_rate": 2.3250410501327037e-09,
+      "loss": 1.58885498046875,
+      "step": 105
+    },
+    {
+      "epoch": 0.351575456053068,
+      "step": 106,
+      "total_flos": 5.062818542541619e+16,
+      "train_loss": 1.7334247456406646,
+      "train_runtime": 876.5295,
+      "train_samples_per_second": 0.482,
+      "train_steps_per_second": 0.121
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 106,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": false,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 5.062818542541619e+16,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}
--- a/training_args.bin
+++ b/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5addc4aa7e357eeb3aee0570e940490d5e41c1b23c3820f34a02dcbe73f67746
+size 6033
--- a/training_loss.png
+++ b/training_loss.png