papiang commited on
Commit
33c7010
·
verified ·
1 Parent(s): 1a1d334

Upload folder using huggingface_hub

Browse files
Files changed (49) hide show
  1. .gitattributes +4 -0
  2. 72/229/README.md +208 -0
  3. 72/229/adapter_config.json +46 -0
  4. 72/229/adapter_model.safetensors +3 -0
  5. 72/229/chat_template.jinja +54 -0
  6. 72/229/optimizer.pt +3 -0
  7. 72/229/rng_state.pth +3 -0
  8. 72/229/scheduler.pt +3 -0
  9. 72/229/tokenizer.json +3 -0
  10. 72/229/tokenizer_config.json +30 -0
  11. 72/229/trainer_state.json +74 -0
  12. 72/229/training_args.bin +3 -0
  13. 72/README.md +60 -0
  14. 72/adapter_config.json +46 -0
  15. 72/adapter_model.safetensors +3 -0
  16. 72/all_results.json +9 -0
  17. 72/chat_template.jinja +54 -0
  18. 72/checkpoint-40/README.md +208 -0
  19. 72/checkpoint-40/adapter_config.json +46 -0
  20. 72/checkpoint-40/adapter_model.safetensors +3 -0
  21. 72/checkpoint-40/chat_template.jinja +54 -0
  22. 72/checkpoint-40/optimizer.pt +3 -0
  23. 72/checkpoint-40/rng_state.pth +3 -0
  24. 72/checkpoint-40/scheduler.pt +3 -0
  25. 72/checkpoint-40/tokenizer.json +3 -0
  26. 72/checkpoint-40/tokenizer_config.json +30 -0
  27. 72/checkpoint-40/trainer_state.json +114 -0
  28. 72/checkpoint-40/training_args.bin +3 -0
  29. 72/checkpoint-60/README.md +208 -0
  30. 72/checkpoint-60/adapter_config.json +46 -0
  31. 72/checkpoint-60/adapter_model.safetensors +3 -0
  32. 72/checkpoint-60/chat_template.jinja +54 -0
  33. 72/checkpoint-60/optimizer.pt +3 -0
  34. 72/checkpoint-60/rng_state.pth +3 -0
  35. 72/checkpoint-60/scheduler.pt +3 -0
  36. 72/checkpoint-60/tokenizer.json +3 -0
  37. 72/checkpoint-60/tokenizer_config.json +30 -0
  38. 72/checkpoint-60/trainer_state.json +154 -0
  39. 72/checkpoint-60/training_args.bin +3 -0
  40. 72/llamaboard_config.yaml +85 -0
  41. 72/running_log.txt +1191 -0
  42. 72/tokenizer.json +3 -0
  43. 72/tokenizer_config.json +30 -0
  44. 72/train_results.json +9 -0
  45. 72/trainer_log.jsonl +13 -0
  46. 72/trainer_state.json +164 -0
  47. 72/training_args.bin +3 -0
  48. 72/training_args.yaml +35 -0
  49. 72/training_loss.png +3 -0
.gitattributes CHANGED
@@ -58,3 +58,7 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
58
  # Video files - compressed
59
  *.mp4 filter=lfs diff=lfs merge=lfs -text
60
  *.webm filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
58
  # Video files - compressed
59
  *.mp4 filter=lfs diff=lfs merge=lfs -text
60
  *.webm filter=lfs diff=lfs merge=lfs -text
61
+ 72/229/tokenizer.json filter=lfs diff=lfs merge=lfs -text
62
+ 72/checkpoint-40/tokenizer.json filter=lfs diff=lfs merge=lfs -text
63
+ 72/checkpoint-60/tokenizer.json filter=lfs diff=lfs merge=lfs -text
64
+ 72/tokenizer.json filter=lfs diff=lfs merge=lfs -text
72/229/README.md ADDED
@@ -0,0 +1,208 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: Qwen/Qwen2.5-72B-Instruct-GPTQ-Int4
3
+ library_name: peft
4
+ pipeline_tag: text-generation
5
+ tags:
6
+ - base_model:adapter:Qwen/Qwen2.5-72B-Instruct-GPTQ-Int4
7
+ - llama-factory
8
+ - lora
9
+ - transformers
10
+ ---
11
+
12
+ # Model Card for Model ID
13
+
14
+ <!-- Provide a quick summary of what the model is/does. -->
15
+
16
+
17
+
18
+ ## Model Details
19
+
20
+ ### Model Description
21
+
22
+ <!-- Provide a longer summary of what this model is. -->
23
+
24
+
25
+
26
+ - **Developed by:** [More Information Needed]
27
+ - **Funded by [optional]:** [More Information Needed]
28
+ - **Shared by [optional]:** [More Information Needed]
29
+ - **Model type:** [More Information Needed]
30
+ - **Language(s) (NLP):** [More Information Needed]
31
+ - **License:** [More Information Needed]
32
+ - **Finetuned from model [optional]:** [More Information Needed]
33
+
34
+ ### Model Sources [optional]
35
+
36
+ <!-- Provide the basic links for the model. -->
37
+
38
+ - **Repository:** [More Information Needed]
39
+ - **Paper [optional]:** [More Information Needed]
40
+ - **Demo [optional]:** [More Information Needed]
41
+
42
+ ## Uses
43
+
44
+ <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
45
+
46
+ ### Direct Use
47
+
48
+ <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
49
+
50
+ [More Information Needed]
51
+
52
+ ### Downstream Use [optional]
53
+
54
+ <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
55
+
56
+ [More Information Needed]
57
+
58
+ ### Out-of-Scope Use
59
+
60
+ <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
61
+
62
+ [More Information Needed]
63
+
64
+ ## Bias, Risks, and Limitations
65
+
66
+ <!-- This section is meant to convey both technical and sociotechnical limitations. -->
67
+
68
+ [More Information Needed]
69
+
70
+ ### Recommendations
71
+
72
+ <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
73
+
74
+ Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
75
+
76
+ ## How to Get Started with the Model
77
+
78
+ Use the code below to get started with the model.
79
+
80
+ [More Information Needed]
81
+
82
+ ## Training Details
83
+
84
+ ### Training Data
85
+
86
+ <!-- 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. -->
87
+
88
+ [More Information Needed]
89
+
90
+ ### Training Procedure
91
+
92
+ <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
93
+
94
+ #### Preprocessing [optional]
95
+
96
+ [More Information Needed]
97
+
98
+
99
+ #### Training Hyperparameters
100
+
101
+ - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
102
+
103
+ #### Speeds, Sizes, Times [optional]
104
+
105
+ <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
106
+
107
+ [More Information Needed]
108
+
109
+ ## Evaluation
110
+
111
+ <!-- This section describes the evaluation protocols and provides the results. -->
112
+
113
+ ### Testing Data, Factors & Metrics
114
+
115
+ #### Testing Data
116
+
117
+ <!-- This should link to a Dataset Card if possible. -->
118
+
119
+ [More Information Needed]
120
+
121
+ #### Factors
122
+
123
+ <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
124
+
125
+ [More Information Needed]
126
+
127
+ #### Metrics
128
+
129
+ <!-- These are the evaluation metrics being used, ideally with a description of why. -->
130
+
131
+ [More Information Needed]
132
+
133
+ ### Results
134
+
135
+ [More Information Needed]
136
+
137
+ #### Summary
138
+
139
+
140
+
141
+ ## Model Examination [optional]
142
+
143
+ <!-- Relevant interpretability work for the model goes here -->
144
+
145
+ [More Information Needed]
146
+
147
+ ## Environmental Impact
148
+
149
+ <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
150
+
151
+ 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).
152
+
153
+ - **Hardware Type:** [More Information Needed]
154
+ - **Hours used:** [More Information Needed]
155
+ - **Cloud Provider:** [More Information Needed]
156
+ - **Compute Region:** [More Information Needed]
157
+ - **Carbon Emitted:** [More Information Needed]
158
+
159
+ ## Technical Specifications [optional]
160
+
161
+ ### Model Architecture and Objective
162
+
163
+ [More Information Needed]
164
+
165
+ ### Compute Infrastructure
166
+
167
+ [More Information Needed]
168
+
169
+ #### Hardware
170
+
171
+ [More Information Needed]
172
+
173
+ #### Software
174
+
175
+ [More Information Needed]
176
+
177
+ ## Citation [optional]
178
+
179
+ <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
180
+
181
+ **BibTeX:**
182
+
183
+ [More Information Needed]
184
+
185
+ **APA:**
186
+
187
+ [More Information Needed]
188
+
189
+ ## Glossary [optional]
190
+
191
+ <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
192
+
193
+ [More Information Needed]
194
+
195
+ ## More Information [optional]
196
+
197
+ [More Information Needed]
198
+
199
+ ## Model Card Authors [optional]
200
+
201
+ [More Information Needed]
202
+
203
+ ## Model Card Contact
204
+
205
+ [More Information Needed]
206
+ ### Framework versions
207
+
208
+ - PEFT 0.18.1
72/229/adapter_config.json ADDED
@@ -0,0 +1,46 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alora_invocation_tokens": null,
3
+ "alpha_pattern": {},
4
+ "arrow_config": null,
5
+ "auto_mapping": null,
6
+ "base_model_name_or_path": "Qwen/Qwen2.5-72B-Instruct-GPTQ-Int4",
7
+ "bias": "none",
8
+ "corda_config": null,
9
+ "ensure_weight_tying": false,
10
+ "eva_config": null,
11
+ "exclude_modules": null,
12
+ "fan_in_fan_out": false,
13
+ "inference_mode": true,
14
+ "init_lora_weights": true,
15
+ "layer_replication": null,
16
+ "layers_pattern": null,
17
+ "layers_to_transform": null,
18
+ "loftq_config": {},
19
+ "lora_alpha": 64,
20
+ "lora_bias": false,
21
+ "lora_dropout": 0.1,
22
+ "megatron_config": null,
23
+ "megatron_core": "megatron.core",
24
+ "modules_to_save": null,
25
+ "peft_type": "LORA",
26
+ "peft_version": "0.18.1",
27
+ "qalora_group_size": 16,
28
+ "r": 32,
29
+ "rank_pattern": {},
30
+ "revision": null,
31
+ "target_modules": [
32
+ "q_proj",
33
+ "v_proj",
34
+ "k_proj",
35
+ "gate_proj",
36
+ "down_proj",
37
+ "up_proj",
38
+ "o_proj"
39
+ ],
40
+ "target_parameters": null,
41
+ "task_type": "CAUSAL_LM",
42
+ "trainable_token_indices": null,
43
+ "use_dora": false,
44
+ "use_qalora": false,
45
+ "use_rslora": false
46
+ }
72/229/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2a971e020adc791218b999c21816558c423845cf5599dd827ddd9f550e0efb97
3
+ size 1688359960
72/229/chat_template.jinja ADDED
@@ -0,0 +1,54 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {%- if tools %}
2
+ {{- '<|im_start|>system\n' }}
3
+ {%- if messages[0]['role'] == 'system' %}
4
+ {{- messages[0]['content'] }}
5
+ {%- else %}
6
+ {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}
7
+ {%- endif %}
8
+ {{- "\n\n# 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>" }}
9
+ {%- for tool in tools %}
10
+ {{- "\n" }}
11
+ {{- tool | tojson }}
12
+ {%- endfor %}
13
+ {{- "\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" }}
14
+ {%- else %}
15
+ {%- if messages[0]['role'] == 'system' %}
16
+ {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
17
+ {%- else %}
18
+ {{- '<|im_start|>system\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\n' }}
19
+ {%- endif %}
20
+ {%- endif %}
21
+ {%- for message in messages %}
22
+ {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
23
+ {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
24
+ {%- elif message.role == "assistant" %}
25
+ {{- '<|im_start|>' + message.role }}
26
+ {%- if message.content %}
27
+ {{- '\n' + message.content }}
28
+ {%- endif %}
29
+ {%- for tool_call in message.tool_calls %}
30
+ {%- if tool_call.function is defined %}
31
+ {%- set tool_call = tool_call.function %}
32
+ {%- endif %}
33
+ {{- '\n<tool_call>\n{"name": "' }}
34
+ {{- tool_call.name }}
35
+ {{- '", "arguments": ' }}
36
+ {{- tool_call.arguments | tojson }}
37
+ {{- '}\n</tool_call>' }}
38
+ {%- endfor %}
39
+ {{- '<|im_end|>\n' }}
40
+ {%- elif message.role == "tool" %}
41
+ {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
42
+ {{- '<|im_start|>user' }}
43
+ {%- endif %}
44
+ {{- '\n<tool_response>\n' }}
45
+ {{- message.content }}
46
+ {{- '\n</tool_response>' }}
47
+ {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
48
+ {{- '<|im_end|>\n' }}
49
+ {%- endif %}
50
+ {%- endif %}
51
+ {%- endfor %}
52
+ {%- if add_generation_prompt %}
53
+ {{- '<|im_start|>assistant\n' }}
54
+ {%- endif %}
72/229/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:81795a2a6f2c1c38c31d2825f96ad13c6c1c0f3ec8bb2949f0986ae013deec02
3
+ size 3377363395
72/229/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:78595db482a95d75737f52c94f17b81e1842fb393e8a496b9bea76d4b07d5877
3
+ size 14645
72/229/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:149a8f0d04002e7e5893c198f21faf729f3c7b5aa2f0a269c71bc13bcb6bcf67
3
+ size 1465
72/229/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3fd169731d2cbde95e10bf356d66d5997fd885dd8dbb6fb4684da3f23b2585d8
3
+ size 11421892
72/229/tokenizer_config.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "backend": "tokenizers",
4
+ "bos_token": null,
5
+ "clean_up_tokenization_spaces": false,
6
+ "eos_token": "<|im_end|>",
7
+ "errors": "replace",
8
+ "extra_special_tokens": [
9
+ "<|im_start|>",
10
+ "<|im_end|>",
11
+ "<|object_ref_start|>",
12
+ "<|object_ref_end|>",
13
+ "<|box_start|>",
14
+ "<|box_end|>",
15
+ "<|quad_start|>",
16
+ "<|quad_end|>",
17
+ "<|vision_start|>",
18
+ "<|vision_end|>",
19
+ "<|vision_pad|>",
20
+ "<|image_pad|>",
21
+ "<|video_pad|>"
22
+ ],
23
+ "is_local": false,
24
+ "model_max_length": 131072,
25
+ "pad_token": "<|endoftext|>",
26
+ "padding_side": "right",
27
+ "split_special_tokens": false,
28
+ "tokenizer_class": "Qwen2Tokenizer",
29
+ "unk_token": null
30
+ }
72/229/trainer_state.json ADDED
@@ -0,0 +1,74 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 5.0,
6
+ "eval_steps": 500,
7
+ "global_step": 20,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 1.2857142857142856,
14
+ "grad_norm": 0.2836259603500366,
15
+ "learning_rate": 4.9692208514878444e-05,
16
+ "loss": 2.557632255554199,
17
+ "num_input_tokens_seen": 346336,
18
+ "step": 5,
19
+ "train_runtime": 1061.9262,
20
+ "train_tokens_per_second": 326.139
21
+ },
22
+ {
23
+ "epoch": 2.571428571428571,
24
+ "grad_norm": 0.20870207250118256,
25
+ "learning_rate": 4.8454783398062106e-05,
26
+ "loss": 2.432936096191406,
27
+ "num_input_tokens_seen": 705016,
28
+ "step": 10,
29
+ "train_runtime": 2156.4566,
30
+ "train_tokens_per_second": 326.933
31
+ },
32
+ {
33
+ "epoch": 3.857142857142857,
34
+ "grad_norm": 0.1494155377149582,
35
+ "learning_rate": 4.6316004108852305e-05,
36
+ "loss": 2.3525007247924803,
37
+ "num_input_tokens_seen": 1041296,
38
+ "step": 15,
39
+ "train_runtime": 3196.4466,
40
+ "train_tokens_per_second": 325.767
41
+ },
42
+ {
43
+ "epoch": 5.0,
44
+ "grad_norm": 0.17206469178199768,
45
+ "learning_rate": 4.335806273589214e-05,
46
+ "loss": 2.2965370178222657,
47
+ "num_input_tokens_seen": 1358800,
48
+ "step": 20,
49
+ "train_runtime": 4165.9384,
50
+ "train_tokens_per_second": 326.169
51
+ }
52
+ ],
53
+ "logging_steps": 5,
54
+ "max_steps": 80,
55
+ "num_input_tokens_seen": 1358800,
56
+ "num_train_epochs": 20,
57
+ "save_steps": 20,
58
+ "stateful_callbacks": {
59
+ "TrainerControl": {
60
+ "args": {
61
+ "should_epoch_stop": false,
62
+ "should_evaluate": false,
63
+ "should_log": false,
64
+ "should_save": true,
65
+ "should_training_stop": false
66
+ },
67
+ "attributes": {}
68
+ }
69
+ },
70
+ "total_flos": 1.3607667597312e+16,
71
+ "train_batch_size": 1,
72
+ "trial_name": null,
73
+ "trial_params": null
74
+ }
72/229/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b70be08810f61028ef84ab71566ed75f3e27e3ed7cd48a3b96c9124bd7cadbbb
3
+ size 5585
72/README.md ADDED
@@ -0,0 +1,60 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ license: other
4
+ base_model: Qwen/Qwen2.5-72B-Instruct-GPTQ-Int4
5
+ tags:
6
+ - base_model:adapter:Qwen/Qwen2.5-72B-Instruct-GPTQ-Int4
7
+ - llama-factory
8
+ - lora
9
+ - transformers
10
+ pipeline_tag: text-generation
11
+ model-index:
12
+ - name: '72'
13
+ results: []
14
+ ---
15
+
16
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
17
+ should probably proofread and complete it, then remove this comment. -->
18
+
19
+ # 72
20
+
21
+ This model is a fine-tuned version of [Qwen/Qwen2.5-72B-Instruct-GPTQ-Int4](https://huggingface.co/Qwen/Qwen2.5-72B-Instruct-GPTQ-Int4) on the historical_sleep_story dataset.
22
+
23
+ ## Model description
24
+
25
+ More information needed
26
+
27
+ ## Intended uses & limitations
28
+
29
+ More information needed
30
+
31
+ ## Training and evaluation data
32
+
33
+ More information needed
34
+
35
+ ## Training procedure
36
+
37
+ ### Training hyperparameters
38
+
39
+ The following hyperparameters were used during training:
40
+ - learning_rate: 5e-05
41
+ - train_batch_size: 1
42
+ - eval_batch_size: 8
43
+ - seed: 42
44
+ - gradient_accumulation_steps: 64
45
+ - total_train_batch_size: 64
46
+ - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
47
+ - lr_scheduler_type: cosine
48
+ - num_epochs: 20.0
49
+
50
+ ### Training results
51
+
52
+
53
+
54
+ ### Framework versions
55
+
56
+ - PEFT 0.18.1
57
+ - Transformers 5.0.0
58
+ - Pytorch 2.10.0+cu128
59
+ - Datasets 4.0.0
60
+ - Tokenizers 0.22.2
72/adapter_config.json ADDED
@@ -0,0 +1,46 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alora_invocation_tokens": null,
3
+ "alpha_pattern": {},
4
+ "arrow_config": null,
5
+ "auto_mapping": null,
6
+ "base_model_name_or_path": "Qwen/Qwen2.5-72B-Instruct-GPTQ-Int4",
7
+ "bias": "none",
8
+ "corda_config": null,
9
+ "ensure_weight_tying": false,
10
+ "eva_config": null,
11
+ "exclude_modules": null,
12
+ "fan_in_fan_out": false,
13
+ "inference_mode": true,
14
+ "init_lora_weights": true,
15
+ "layer_replication": null,
16
+ "layers_pattern": null,
17
+ "layers_to_transform": null,
18
+ "loftq_config": {},
19
+ "lora_alpha": 64,
20
+ "lora_bias": false,
21
+ "lora_dropout": 0.1,
22
+ "megatron_config": null,
23
+ "megatron_core": "megatron.core",
24
+ "modules_to_save": null,
25
+ "peft_type": "LORA",
26
+ "peft_version": "0.18.1",
27
+ "qalora_group_size": 16,
28
+ "r": 32,
29
+ "rank_pattern": {},
30
+ "revision": null,
31
+ "target_modules": [
32
+ "q_proj",
33
+ "v_proj",
34
+ "k_proj",
35
+ "gate_proj",
36
+ "down_proj",
37
+ "up_proj",
38
+ "o_proj"
39
+ ],
40
+ "target_parameters": null,
41
+ "task_type": "CAUSAL_LM",
42
+ "trainable_token_indices": null,
43
+ "use_dora": false,
44
+ "use_qalora": false,
45
+ "use_rslora": false
46
+ }
72/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f317dae4a8498ce6c741f6505432db43247a0341a55ab732abd11e6284dea0d0
3
+ size 1688359960
72/all_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 15.0,
3
+ "num_input_tokens_seen": 4085560,
4
+ "total_flos": 4.09147353759744e+16,
5
+ "train_loss": 2.195078119387229,
6
+ "train_runtime": 12540.6077,
7
+ "train_samples_per_second": 0.357,
8
+ "train_steps_per_second": 0.006
9
+ }
72/chat_template.jinja ADDED
@@ -0,0 +1,54 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {%- if tools %}
2
+ {{- '<|im_start|>system\n' }}
3
+ {%- if messages[0]['role'] == 'system' %}
4
+ {{- messages[0]['content'] }}
5
+ {%- else %}
6
+ {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}
7
+ {%- endif %}
8
+ {{- "\n\n# 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>" }}
9
+ {%- for tool in tools %}
10
+ {{- "\n" }}
11
+ {{- tool | tojson }}
12
+ {%- endfor %}
13
+ {{- "\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" }}
14
+ {%- else %}
15
+ {%- if messages[0]['role'] == 'system' %}
16
+ {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
17
+ {%- else %}
18
+ {{- '<|im_start|>system\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\n' }}
19
+ {%- endif %}
20
+ {%- endif %}
21
+ {%- for message in messages %}
22
+ {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
23
+ {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
24
+ {%- elif message.role == "assistant" %}
25
+ {{- '<|im_start|>' + message.role }}
26
+ {%- if message.content %}
27
+ {{- '\n' + message.content }}
28
+ {%- endif %}
29
+ {%- for tool_call in message.tool_calls %}
30
+ {%- if tool_call.function is defined %}
31
+ {%- set tool_call = tool_call.function %}
32
+ {%- endif %}
33
+ {{- '\n<tool_call>\n{"name": "' }}
34
+ {{- tool_call.name }}
35
+ {{- '", "arguments": ' }}
36
+ {{- tool_call.arguments | tojson }}
37
+ {{- '}\n</tool_call>' }}
38
+ {%- endfor %}
39
+ {{- '<|im_end|>\n' }}
40
+ {%- elif message.role == "tool" %}
41
+ {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
42
+ {{- '<|im_start|>user' }}
43
+ {%- endif %}
44
+ {{- '\n<tool_response>\n' }}
45
+ {{- message.content }}
46
+ {{- '\n</tool_response>' }}
47
+ {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
48
+ {{- '<|im_end|>\n' }}
49
+ {%- endif %}
50
+ {%- endif %}
51
+ {%- endfor %}
52
+ {%- if add_generation_prompt %}
53
+ {{- '<|im_start|>assistant\n' }}
54
+ {%- endif %}
72/checkpoint-40/README.md ADDED
@@ -0,0 +1,208 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: Qwen/Qwen2.5-72B-Instruct-GPTQ-Int4
3
+ library_name: peft
4
+ pipeline_tag: text-generation
5
+ tags:
6
+ - base_model:adapter:Qwen/Qwen2.5-72B-Instruct-GPTQ-Int4
7
+ - llama-factory
8
+ - lora
9
+ - transformers
10
+ ---
11
+
12
+ # Model Card for Model ID
13
+
14
+ <!-- Provide a quick summary of what the model is/does. -->
15
+
16
+
17
+
18
+ ## Model Details
19
+
20
+ ### Model Description
21
+
22
+ <!-- Provide a longer summary of what this model is. -->
23
+
24
+
25
+
26
+ - **Developed by:** [More Information Needed]
27
+ - **Funded by [optional]:** [More Information Needed]
28
+ - **Shared by [optional]:** [More Information Needed]
29
+ - **Model type:** [More Information Needed]
30
+ - **Language(s) (NLP):** [More Information Needed]
31
+ - **License:** [More Information Needed]
32
+ - **Finetuned from model [optional]:** [More Information Needed]
33
+
34
+ ### Model Sources [optional]
35
+
36
+ <!-- Provide the basic links for the model. -->
37
+
38
+ - **Repository:** [More Information Needed]
39
+ - **Paper [optional]:** [More Information Needed]
40
+ - **Demo [optional]:** [More Information Needed]
41
+
42
+ ## Uses
43
+
44
+ <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
45
+
46
+ ### Direct Use
47
+
48
+ <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
49
+
50
+ [More Information Needed]
51
+
52
+ ### Downstream Use [optional]
53
+
54
+ <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
55
+
56
+ [More Information Needed]
57
+
58
+ ### Out-of-Scope Use
59
+
60
+ <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
61
+
62
+ [More Information Needed]
63
+
64
+ ## Bias, Risks, and Limitations
65
+
66
+ <!-- This section is meant to convey both technical and sociotechnical limitations. -->
67
+
68
+ [More Information Needed]
69
+
70
+ ### Recommendations
71
+
72
+ <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
73
+
74
+ Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
75
+
76
+ ## How to Get Started with the Model
77
+
78
+ Use the code below to get started with the model.
79
+
80
+ [More Information Needed]
81
+
82
+ ## Training Details
83
+
84
+ ### Training Data
85
+
86
+ <!-- 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. -->
87
+
88
+ [More Information Needed]
89
+
90
+ ### Training Procedure
91
+
92
+ <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
93
+
94
+ #### Preprocessing [optional]
95
+
96
+ [More Information Needed]
97
+
98
+
99
+ #### Training Hyperparameters
100
+
101
+ - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
102
+
103
+ #### Speeds, Sizes, Times [optional]
104
+
105
+ <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
106
+
107
+ [More Information Needed]
108
+
109
+ ## Evaluation
110
+
111
+ <!-- This section describes the evaluation protocols and provides the results. -->
112
+
113
+ ### Testing Data, Factors & Metrics
114
+
115
+ #### Testing Data
116
+
117
+ <!-- This should link to a Dataset Card if possible. -->
118
+
119
+ [More Information Needed]
120
+
121
+ #### Factors
122
+
123
+ <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
124
+
125
+ [More Information Needed]
126
+
127
+ #### Metrics
128
+
129
+ <!-- These are the evaluation metrics being used, ideally with a description of why. -->
130
+
131
+ [More Information Needed]
132
+
133
+ ### Results
134
+
135
+ [More Information Needed]
136
+
137
+ #### Summary
138
+
139
+
140
+
141
+ ## Model Examination [optional]
142
+
143
+ <!-- Relevant interpretability work for the model goes here -->
144
+
145
+ [More Information Needed]
146
+
147
+ ## Environmental Impact
148
+
149
+ <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
150
+
151
+ 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).
152
+
153
+ - **Hardware Type:** [More Information Needed]
154
+ - **Hours used:** [More Information Needed]
155
+ - **Cloud Provider:** [More Information Needed]
156
+ - **Compute Region:** [More Information Needed]
157
+ - **Carbon Emitted:** [More Information Needed]
158
+
159
+ ## Technical Specifications [optional]
160
+
161
+ ### Model Architecture and Objective
162
+
163
+ [More Information Needed]
164
+
165
+ ### Compute Infrastructure
166
+
167
+ [More Information Needed]
168
+
169
+ #### Hardware
170
+
171
+ [More Information Needed]
172
+
173
+ #### Software
174
+
175
+ [More Information Needed]
176
+
177
+ ## Citation [optional]
178
+
179
+ <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
180
+
181
+ **BibTeX:**
182
+
183
+ [More Information Needed]
184
+
185
+ **APA:**
186
+
187
+ [More Information Needed]
188
+
189
+ ## Glossary [optional]
190
+
191
+ <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
192
+
193
+ [More Information Needed]
194
+
195
+ ## More Information [optional]
196
+
197
+ [More Information Needed]
198
+
199
+ ## Model Card Authors [optional]
200
+
201
+ [More Information Needed]
202
+
203
+ ## Model Card Contact
204
+
205
+ [More Information Needed]
206
+ ### Framework versions
207
+
208
+ - PEFT 0.18.1
72/checkpoint-40/adapter_config.json ADDED
@@ -0,0 +1,46 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alora_invocation_tokens": null,
3
+ "alpha_pattern": {},
4
+ "arrow_config": null,
5
+ "auto_mapping": null,
6
+ "base_model_name_or_path": "Qwen/Qwen2.5-72B-Instruct-GPTQ-Int4",
7
+ "bias": "none",
8
+ "corda_config": null,
9
+ "ensure_weight_tying": false,
10
+ "eva_config": null,
11
+ "exclude_modules": null,
12
+ "fan_in_fan_out": false,
13
+ "inference_mode": true,
14
+ "init_lora_weights": true,
15
+ "layer_replication": null,
16
+ "layers_pattern": null,
17
+ "layers_to_transform": null,
18
+ "loftq_config": {},
19
+ "lora_alpha": 64,
20
+ "lora_bias": false,
21
+ "lora_dropout": 0.1,
22
+ "megatron_config": null,
23
+ "megatron_core": "megatron.core",
24
+ "modules_to_save": null,
25
+ "peft_type": "LORA",
26
+ "peft_version": "0.18.1",
27
+ "qalora_group_size": 16,
28
+ "r": 32,
29
+ "rank_pattern": {},
30
+ "revision": null,
31
+ "target_modules": [
32
+ "q_proj",
33
+ "v_proj",
34
+ "k_proj",
35
+ "gate_proj",
36
+ "down_proj",
37
+ "up_proj",
38
+ "o_proj"
39
+ ],
40
+ "target_parameters": null,
41
+ "task_type": "CAUSAL_LM",
42
+ "trainable_token_indices": null,
43
+ "use_dora": false,
44
+ "use_qalora": false,
45
+ "use_rslora": false
46
+ }
72/checkpoint-40/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cf3dee3d365107d2aeec43c82abc348d74564b2e2896bea58a4de89277bd2d6d
3
+ size 1688359960
72/checkpoint-40/chat_template.jinja ADDED
@@ -0,0 +1,54 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {%- if tools %}
2
+ {{- '<|im_start|>system\n' }}
3
+ {%- if messages[0]['role'] == 'system' %}
4
+ {{- messages[0]['content'] }}
5
+ {%- else %}
6
+ {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}
7
+ {%- endif %}
8
+ {{- "\n\n# 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>" }}
9
+ {%- for tool in tools %}
10
+ {{- "\n" }}
11
+ {{- tool | tojson }}
12
+ {%- endfor %}
13
+ {{- "\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" }}
14
+ {%- else %}
15
+ {%- if messages[0]['role'] == 'system' %}
16
+ {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
17
+ {%- else %}
18
+ {{- '<|im_start|>system\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\n' }}
19
+ {%- endif %}
20
+ {%- endif %}
21
+ {%- for message in messages %}
22
+ {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
23
+ {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
24
+ {%- elif message.role == "assistant" %}
25
+ {{- '<|im_start|>' + message.role }}
26
+ {%- if message.content %}
27
+ {{- '\n' + message.content }}
28
+ {%- endif %}
29
+ {%- for tool_call in message.tool_calls %}
30
+ {%- if tool_call.function is defined %}
31
+ {%- set tool_call = tool_call.function %}
32
+ {%- endif %}
33
+ {{- '\n<tool_call>\n{"name": "' }}
34
+ {{- tool_call.name }}
35
+ {{- '", "arguments": ' }}
36
+ {{- tool_call.arguments | tojson }}
37
+ {{- '}\n</tool_call>' }}
38
+ {%- endfor %}
39
+ {{- '<|im_end|>\n' }}
40
+ {%- elif message.role == "tool" %}
41
+ {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
42
+ {{- '<|im_start|>user' }}
43
+ {%- endif %}
44
+ {{- '\n<tool_response>\n' }}
45
+ {{- message.content }}
46
+ {{- '\n</tool_response>' }}
47
+ {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
48
+ {{- '<|im_end|>\n' }}
49
+ {%- endif %}
50
+ {%- endif %}
51
+ {%- endfor %}
52
+ {%- if add_generation_prompt %}
53
+ {{- '<|im_start|>assistant\n' }}
54
+ {%- endif %}
72/checkpoint-40/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6c6c57a86a092ac6775d63d61c0b533a8676ad07d4b6fb214a4dadc765e41554
3
+ size 3377363395
72/checkpoint-40/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fba3561c3ec15002364a6be1cb991800f7f6c51e5dd7a6a7b01dbc59dd46e9ed
3
+ size 14645
72/checkpoint-40/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5c07693346cd2d28a5391feee426ba3efb7c029a9c1d9cce1cbfe76ce4f6cbc5
3
+ size 1465
72/checkpoint-40/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3fd169731d2cbde95e10bf356d66d5997fd885dd8dbb6fb4684da3f23b2585d8
3
+ size 11421892
72/checkpoint-40/tokenizer_config.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "backend": "tokenizers",
4
+ "bos_token": null,
5
+ "clean_up_tokenization_spaces": false,
6
+ "eos_token": "<|im_end|>",
7
+ "errors": "replace",
8
+ "extra_special_tokens": [
9
+ "<|im_start|>",
10
+ "<|im_end|>",
11
+ "<|object_ref_start|>",
12
+ "<|object_ref_end|>",
13
+ "<|box_start|>",
14
+ "<|box_end|>",
15
+ "<|quad_start|>",
16
+ "<|quad_end|>",
17
+ "<|vision_start|>",
18
+ "<|vision_end|>",
19
+ "<|vision_pad|>",
20
+ "<|image_pad|>",
21
+ "<|video_pad|>"
22
+ ],
23
+ "is_local": false,
24
+ "model_max_length": 131072,
25
+ "pad_token": "<|endoftext|>",
26
+ "padding_side": "right",
27
+ "split_special_tokens": false,
28
+ "tokenizer_class": "Qwen2Tokenizer",
29
+ "unk_token": null
30
+ }
72/checkpoint-40/trainer_state.json ADDED
@@ -0,0 +1,114 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 10.0,
6
+ "eval_steps": 500,
7
+ "global_step": 40,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 1.2857142857142856,
14
+ "grad_norm": 0.2836259603500366,
15
+ "learning_rate": 4.9692208514878444e-05,
16
+ "loss": 2.557632255554199,
17
+ "num_input_tokens_seen": 346336,
18
+ "step": 5,
19
+ "train_runtime": 1061.9262,
20
+ "train_tokens_per_second": 326.139
21
+ },
22
+ {
23
+ "epoch": 2.571428571428571,
24
+ "grad_norm": 0.20870207250118256,
25
+ "learning_rate": 4.8454783398062106e-05,
26
+ "loss": 2.432936096191406,
27
+ "num_input_tokens_seen": 705016,
28
+ "step": 10,
29
+ "train_runtime": 2156.4566,
30
+ "train_tokens_per_second": 326.933
31
+ },
32
+ {
33
+ "epoch": 3.857142857142857,
34
+ "grad_norm": 0.1494155377149582,
35
+ "learning_rate": 4.6316004108852305e-05,
36
+ "loss": 2.3525007247924803,
37
+ "num_input_tokens_seen": 1041296,
38
+ "step": 15,
39
+ "train_runtime": 3196.4466,
40
+ "train_tokens_per_second": 325.767
41
+ },
42
+ {
43
+ "epoch": 5.0,
44
+ "grad_norm": 0.17206469178199768,
45
+ "learning_rate": 4.335806273589214e-05,
46
+ "loss": 2.2965370178222657,
47
+ "num_input_tokens_seen": 1358800,
48
+ "step": 20,
49
+ "train_runtime": 4165.9384,
50
+ "train_tokens_per_second": 326.169
51
+ },
52
+ {
53
+ "epoch": 6.285714285714286,
54
+ "grad_norm": 0.14012837409973145,
55
+ "learning_rate": 3.969463130731183e-05,
56
+ "loss": 2.2461246490478515,
57
+ "num_input_tokens_seen": 1703832,
58
+ "step": 25,
59
+ "train_runtime": 5229.5525,
60
+ "train_tokens_per_second": 325.808
61
+ },
62
+ {
63
+ "epoch": 7.571428571428571,
64
+ "grad_norm": 0.15643367171287537,
65
+ "learning_rate": 3.54664934384357e-05,
66
+ "loss": 2.1892810821533204,
67
+ "num_input_tokens_seen": 2052472,
68
+ "step": 30,
69
+ "train_runtime": 6300.2335,
70
+ "train_tokens_per_second": 325.777
71
+ },
72
+ {
73
+ "epoch": 8.857142857142858,
74
+ "grad_norm": 0.1400991678237915,
75
+ "learning_rate": 3.083613409639764e-05,
76
+ "loss": 2.1421274185180663,
77
+ "num_input_tokens_seen": 2412408,
78
+ "step": 35,
79
+ "train_runtime": 7395.9727,
80
+ "train_tokens_per_second": 326.179
81
+ },
82
+ {
83
+ "epoch": 10.0,
84
+ "grad_norm": 0.24552510678768158,
85
+ "learning_rate": 2.598149539397672e-05,
86
+ "loss": 2.097066307067871,
87
+ "num_input_tokens_seen": 2717600,
88
+ "step": 40,
89
+ "train_runtime": 8336.2619,
90
+ "train_tokens_per_second": 325.997
91
+ }
92
+ ],
93
+ "logging_steps": 5,
94
+ "max_steps": 80,
95
+ "num_input_tokens_seen": 2717600,
96
+ "num_train_epochs": 20,
97
+ "save_steps": 20,
98
+ "stateful_callbacks": {
99
+ "TrainerControl": {
100
+ "args": {
101
+ "should_epoch_stop": false,
102
+ "should_evaluate": false,
103
+ "should_log": false,
104
+ "should_save": true,
105
+ "should_training_stop": false
106
+ },
107
+ "attributes": {}
108
+ }
109
+ },
110
+ "total_flos": 2.7215335194624e+16,
111
+ "train_batch_size": 1,
112
+ "trial_name": null,
113
+ "trial_params": null
114
+ }
72/checkpoint-40/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b70be08810f61028ef84ab71566ed75f3e27e3ed7cd48a3b96c9124bd7cadbbb
3
+ size 5585
72/checkpoint-60/README.md ADDED
@@ -0,0 +1,208 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: Qwen/Qwen2.5-72B-Instruct-GPTQ-Int4
3
+ library_name: peft
4
+ pipeline_tag: text-generation
5
+ tags:
6
+ - base_model:adapter:Qwen/Qwen2.5-72B-Instruct-GPTQ-Int4
7
+ - llama-factory
8
+ - lora
9
+ - transformers
10
+ ---
11
+
12
+ # Model Card for Model ID
13
+
14
+ <!-- Provide a quick summary of what the model is/does. -->
15
+
16
+
17
+
18
+ ## Model Details
19
+
20
+ ### Model Description
21
+
22
+ <!-- Provide a longer summary of what this model is. -->
23
+
24
+
25
+
26
+ - **Developed by:** [More Information Needed]
27
+ - **Funded by [optional]:** [More Information Needed]
28
+ - **Shared by [optional]:** [More Information Needed]
29
+ - **Model type:** [More Information Needed]
30
+ - **Language(s) (NLP):** [More Information Needed]
31
+ - **License:** [More Information Needed]
32
+ - **Finetuned from model [optional]:** [More Information Needed]
33
+
34
+ ### Model Sources [optional]
35
+
36
+ <!-- Provide the basic links for the model. -->
37
+
38
+ - **Repository:** [More Information Needed]
39
+ - **Paper [optional]:** [More Information Needed]
40
+ - **Demo [optional]:** [More Information Needed]
41
+
42
+ ## Uses
43
+
44
+ <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
45
+
46
+ ### Direct Use
47
+
48
+ <!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
49
+
50
+ [More Information Needed]
51
+
52
+ ### Downstream Use [optional]
53
+
54
+ <!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
55
+
56
+ [More Information Needed]
57
+
58
+ ### Out-of-Scope Use
59
+
60
+ <!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
61
+
62
+ [More Information Needed]
63
+
64
+ ## Bias, Risks, and Limitations
65
+
66
+ <!-- This section is meant to convey both technical and sociotechnical limitations. -->
67
+
68
+ [More Information Needed]
69
+
70
+ ### Recommendations
71
+
72
+ <!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
73
+
74
+ Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
75
+
76
+ ## How to Get Started with the Model
77
+
78
+ Use the code below to get started with the model.
79
+
80
+ [More Information Needed]
81
+
82
+ ## Training Details
83
+
84
+ ### Training Data
85
+
86
+ <!-- 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. -->
87
+
88
+ [More Information Needed]
89
+
90
+ ### Training Procedure
91
+
92
+ <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
93
+
94
+ #### Preprocessing [optional]
95
+
96
+ [More Information Needed]
97
+
98
+
99
+ #### Training Hyperparameters
100
+
101
+ - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
102
+
103
+ #### Speeds, Sizes, Times [optional]
104
+
105
+ <!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
106
+
107
+ [More Information Needed]
108
+
109
+ ## Evaluation
110
+
111
+ <!-- This section describes the evaluation protocols and provides the results. -->
112
+
113
+ ### Testing Data, Factors & Metrics
114
+
115
+ #### Testing Data
116
+
117
+ <!-- This should link to a Dataset Card if possible. -->
118
+
119
+ [More Information Needed]
120
+
121
+ #### Factors
122
+
123
+ <!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
124
+
125
+ [More Information Needed]
126
+
127
+ #### Metrics
128
+
129
+ <!-- These are the evaluation metrics being used, ideally with a description of why. -->
130
+
131
+ [More Information Needed]
132
+
133
+ ### Results
134
+
135
+ [More Information Needed]
136
+
137
+ #### Summary
138
+
139
+
140
+
141
+ ## Model Examination [optional]
142
+
143
+ <!-- Relevant interpretability work for the model goes here -->
144
+
145
+ [More Information Needed]
146
+
147
+ ## Environmental Impact
148
+
149
+ <!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
150
+
151
+ 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).
152
+
153
+ - **Hardware Type:** [More Information Needed]
154
+ - **Hours used:** [More Information Needed]
155
+ - **Cloud Provider:** [More Information Needed]
156
+ - **Compute Region:** [More Information Needed]
157
+ - **Carbon Emitted:** [More Information Needed]
158
+
159
+ ## Technical Specifications [optional]
160
+
161
+ ### Model Architecture and Objective
162
+
163
+ [More Information Needed]
164
+
165
+ ### Compute Infrastructure
166
+
167
+ [More Information Needed]
168
+
169
+ #### Hardware
170
+
171
+ [More Information Needed]
172
+
173
+ #### Software
174
+
175
+ [More Information Needed]
176
+
177
+ ## Citation [optional]
178
+
179
+ <!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
180
+
181
+ **BibTeX:**
182
+
183
+ [More Information Needed]
184
+
185
+ **APA:**
186
+
187
+ [More Information Needed]
188
+
189
+ ## Glossary [optional]
190
+
191
+ <!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
192
+
193
+ [More Information Needed]
194
+
195
+ ## More Information [optional]
196
+
197
+ [More Information Needed]
198
+
199
+ ## Model Card Authors [optional]
200
+
201
+ [More Information Needed]
202
+
203
+ ## Model Card Contact
204
+
205
+ [More Information Needed]
206
+ ### Framework versions
207
+
208
+ - PEFT 0.18.1
72/checkpoint-60/adapter_config.json ADDED
@@ -0,0 +1,46 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alora_invocation_tokens": null,
3
+ "alpha_pattern": {},
4
+ "arrow_config": null,
5
+ "auto_mapping": null,
6
+ "base_model_name_or_path": "Qwen/Qwen2.5-72B-Instruct-GPTQ-Int4",
7
+ "bias": "none",
8
+ "corda_config": null,
9
+ "ensure_weight_tying": false,
10
+ "eva_config": null,
11
+ "exclude_modules": null,
12
+ "fan_in_fan_out": false,
13
+ "inference_mode": true,
14
+ "init_lora_weights": true,
15
+ "layer_replication": null,
16
+ "layers_pattern": null,
17
+ "layers_to_transform": null,
18
+ "loftq_config": {},
19
+ "lora_alpha": 64,
20
+ "lora_bias": false,
21
+ "lora_dropout": 0.1,
22
+ "megatron_config": null,
23
+ "megatron_core": "megatron.core",
24
+ "modules_to_save": null,
25
+ "peft_type": "LORA",
26
+ "peft_version": "0.18.1",
27
+ "qalora_group_size": 16,
28
+ "r": 32,
29
+ "rank_pattern": {},
30
+ "revision": null,
31
+ "target_modules": [
32
+ "q_proj",
33
+ "v_proj",
34
+ "k_proj",
35
+ "gate_proj",
36
+ "down_proj",
37
+ "up_proj",
38
+ "o_proj"
39
+ ],
40
+ "target_parameters": null,
41
+ "task_type": "CAUSAL_LM",
42
+ "trainable_token_indices": null,
43
+ "use_dora": false,
44
+ "use_qalora": false,
45
+ "use_rslora": false
46
+ }
72/checkpoint-60/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f317dae4a8498ce6c741f6505432db43247a0341a55ab732abd11e6284dea0d0
3
+ size 1688359960
72/checkpoint-60/chat_template.jinja ADDED
@@ -0,0 +1,54 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {%- if tools %}
2
+ {{- '<|im_start|>system\n' }}
3
+ {%- if messages[0]['role'] == 'system' %}
4
+ {{- messages[0]['content'] }}
5
+ {%- else %}
6
+ {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}
7
+ {%- endif %}
8
+ {{- "\n\n# 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>" }}
9
+ {%- for tool in tools %}
10
+ {{- "\n" }}
11
+ {{- tool | tojson }}
12
+ {%- endfor %}
13
+ {{- "\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" }}
14
+ {%- else %}
15
+ {%- if messages[0]['role'] == 'system' %}
16
+ {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
17
+ {%- else %}
18
+ {{- '<|im_start|>system\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\n' }}
19
+ {%- endif %}
20
+ {%- endif %}
21
+ {%- for message in messages %}
22
+ {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
23
+ {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
24
+ {%- elif message.role == "assistant" %}
25
+ {{- '<|im_start|>' + message.role }}
26
+ {%- if message.content %}
27
+ {{- '\n' + message.content }}
28
+ {%- endif %}
29
+ {%- for tool_call in message.tool_calls %}
30
+ {%- if tool_call.function is defined %}
31
+ {%- set tool_call = tool_call.function %}
32
+ {%- endif %}
33
+ {{- '\n<tool_call>\n{"name": "' }}
34
+ {{- tool_call.name }}
35
+ {{- '", "arguments": ' }}
36
+ {{- tool_call.arguments | tojson }}
37
+ {{- '}\n</tool_call>' }}
38
+ {%- endfor %}
39
+ {{- '<|im_end|>\n' }}
40
+ {%- elif message.role == "tool" %}
41
+ {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
42
+ {{- '<|im_start|>user' }}
43
+ {%- endif %}
44
+ {{- '\n<tool_response>\n' }}
45
+ {{- message.content }}
46
+ {{- '\n</tool_response>' }}
47
+ {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
48
+ {{- '<|im_end|>\n' }}
49
+ {%- endif %}
50
+ {%- endif %}
51
+ {%- endfor %}
52
+ {%- if add_generation_prompt %}
53
+ {{- '<|im_start|>assistant\n' }}
54
+ {%- endif %}
72/checkpoint-60/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5ddd7b87ce15f00cb180447408d6bc482a371485dbb41c2602a48f18e859f042
3
+ size 3377363395
72/checkpoint-60/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:11dbacd87dd479df99e2b4b3797197d5708c30c4d6490de4c1aae6368aa9762f
3
+ size 14645
72/checkpoint-60/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c47493b2311f660106d74553f978fe6027d37f161b9fe57e7dd7490508d35415
3
+ size 1465
72/checkpoint-60/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3fd169731d2cbde95e10bf356d66d5997fd885dd8dbb6fb4684da3f23b2585d8
3
+ size 11421892
72/checkpoint-60/tokenizer_config.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "backend": "tokenizers",
4
+ "bos_token": null,
5
+ "clean_up_tokenization_spaces": false,
6
+ "eos_token": "<|im_end|>",
7
+ "errors": "replace",
8
+ "extra_special_tokens": [
9
+ "<|im_start|>",
10
+ "<|im_end|>",
11
+ "<|object_ref_start|>",
12
+ "<|object_ref_end|>",
13
+ "<|box_start|>",
14
+ "<|box_end|>",
15
+ "<|quad_start|>",
16
+ "<|quad_end|>",
17
+ "<|vision_start|>",
18
+ "<|vision_end|>",
19
+ "<|vision_pad|>",
20
+ "<|image_pad|>",
21
+ "<|video_pad|>"
22
+ ],
23
+ "is_local": false,
24
+ "model_max_length": 131072,
25
+ "pad_token": "<|endoftext|>",
26
+ "padding_side": "right",
27
+ "split_special_tokens": false,
28
+ "tokenizer_class": "Qwen2Tokenizer",
29
+ "unk_token": null
30
+ }
72/checkpoint-60/trainer_state.json ADDED
@@ -0,0 +1,154 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 15.0,
6
+ "eval_steps": 500,
7
+ "global_step": 60,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 1.2857142857142856,
14
+ "grad_norm": 0.2836259603500366,
15
+ "learning_rate": 4.9692208514878444e-05,
16
+ "loss": 2.557632255554199,
17
+ "num_input_tokens_seen": 346336,
18
+ "step": 5,
19
+ "train_runtime": 1061.9262,
20
+ "train_tokens_per_second": 326.139
21
+ },
22
+ {
23
+ "epoch": 2.571428571428571,
24
+ "grad_norm": 0.20870207250118256,
25
+ "learning_rate": 4.8454783398062106e-05,
26
+ "loss": 2.432936096191406,
27
+ "num_input_tokens_seen": 705016,
28
+ "step": 10,
29
+ "train_runtime": 2156.4566,
30
+ "train_tokens_per_second": 326.933
31
+ },
32
+ {
33
+ "epoch": 3.857142857142857,
34
+ "grad_norm": 0.1494155377149582,
35
+ "learning_rate": 4.6316004108852305e-05,
36
+ "loss": 2.3525007247924803,
37
+ "num_input_tokens_seen": 1041296,
38
+ "step": 15,
39
+ "train_runtime": 3196.4466,
40
+ "train_tokens_per_second": 325.767
41
+ },
42
+ {
43
+ "epoch": 5.0,
44
+ "grad_norm": 0.17206469178199768,
45
+ "learning_rate": 4.335806273589214e-05,
46
+ "loss": 2.2965370178222657,
47
+ "num_input_tokens_seen": 1358800,
48
+ "step": 20,
49
+ "train_runtime": 4165.9384,
50
+ "train_tokens_per_second": 326.169
51
+ },
52
+ {
53
+ "epoch": 6.285714285714286,
54
+ "grad_norm": 0.14012837409973145,
55
+ "learning_rate": 3.969463130731183e-05,
56
+ "loss": 2.2461246490478515,
57
+ "num_input_tokens_seen": 1703832,
58
+ "step": 25,
59
+ "train_runtime": 5229.5525,
60
+ "train_tokens_per_second": 325.808
61
+ },
62
+ {
63
+ "epoch": 7.571428571428571,
64
+ "grad_norm": 0.15643367171287537,
65
+ "learning_rate": 3.54664934384357e-05,
66
+ "loss": 2.1892810821533204,
67
+ "num_input_tokens_seen": 2052472,
68
+ "step": 30,
69
+ "train_runtime": 6300.2335,
70
+ "train_tokens_per_second": 325.777
71
+ },
72
+ {
73
+ "epoch": 8.857142857142858,
74
+ "grad_norm": 0.1400991678237915,
75
+ "learning_rate": 3.083613409639764e-05,
76
+ "loss": 2.1421274185180663,
77
+ "num_input_tokens_seen": 2412408,
78
+ "step": 35,
79
+ "train_runtime": 7395.9727,
80
+ "train_tokens_per_second": 326.179
81
+ },
82
+ {
83
+ "epoch": 10.0,
84
+ "grad_norm": 0.24552510678768158,
85
+ "learning_rate": 2.598149539397672e-05,
86
+ "loss": 2.097066307067871,
87
+ "num_input_tokens_seen": 2717600,
88
+ "step": 40,
89
+ "train_runtime": 8336.2619,
90
+ "train_tokens_per_second": 325.997
91
+ },
92
+ {
93
+ "epoch": 11.285714285714286,
94
+ "grad_norm": 0.17399123311042786,
95
+ "learning_rate": 2.1089138373994223e-05,
96
+ "loss": 2.046438980102539,
97
+ "num_input_tokens_seen": 3064544,
98
+ "step": 45,
99
+ "train_runtime": 9405.8661,
100
+ "train_tokens_per_second": 325.812
101
+ },
102
+ {
103
+ "epoch": 12.571428571428571,
104
+ "grad_norm": 0.19238384068012238,
105
+ "learning_rate": 1.6347073573062672e-05,
106
+ "loss": 2.0126678466796877,
107
+ "num_input_tokens_seen": 3421576,
108
+ "step": 50,
109
+ "train_runtime": 10500.9313,
110
+ "train_tokens_per_second": 325.835
111
+ },
112
+ {
113
+ "epoch": 13.857142857142858,
114
+ "grad_norm": 0.2132018357515335,
115
+ "learning_rate": 1.1937535882101281e-05,
116
+ "loss": 1.9826169967651368,
117
+ "num_input_tokens_seen": 3766296,
118
+ "step": 55,
119
+ "train_runtime": 11560.8898,
120
+ "train_tokens_per_second": 325.779
121
+ },
122
+ {
123
+ "epoch": 15.0,
124
+ "grad_norm": 0.3054792881011963,
125
+ "learning_rate": 8.029981361676456e-06,
126
+ "loss": 1.9407329559326172,
127
+ "num_input_tokens_seen": 4076400,
128
+ "step": 60,
129
+ "train_runtime": 12513.0566,
130
+ "train_tokens_per_second": 325.772
131
+ }
132
+ ],
133
+ "logging_steps": 5,
134
+ "max_steps": 80,
135
+ "num_input_tokens_seen": 4076400,
136
+ "num_train_epochs": 20,
137
+ "save_steps": 20,
138
+ "stateful_callbacks": {
139
+ "TrainerControl": {
140
+ "args": {
141
+ "should_epoch_stop": false,
142
+ "should_evaluate": false,
143
+ "should_log": false,
144
+ "should_save": true,
145
+ "should_training_stop": false
146
+ },
147
+ "attributes": {}
148
+ }
149
+ },
150
+ "total_flos": 4.0823002791936e+16,
151
+ "train_batch_size": 1,
152
+ "trial_name": null,
153
+ "trial_params": null
154
+ }
72/checkpoint-60/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b70be08810f61028ef84ab71566ed75f3e27e3ed7cd48a3b96c9124bd7cadbbb
3
+ size 5585
72/llamaboard_config.yaml ADDED
@@ -0,0 +1,85 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ top.booster: auto
2
+ top.checkpoint_path: []
3
+ top.finetuning_type: lora
4
+ top.model_name: Qwen2.5-72B-Instruct-GPTQ-Int4
5
+ top.quantization_bit: none
6
+ top.quantization_method: bnb
7
+ top.rope_scaling: none
8
+ top.template: qwen
9
+ train.additional_target: ''
10
+ train.apollo_rank: 16
11
+ train.apollo_scale: 32
12
+ train.apollo_target: all
13
+ train.apollo_update_interval: 200
14
+ train.badam_mode: layer
15
+ train.badam_switch_interval: 50
16
+ train.badam_switch_mode: ascending
17
+ train.badam_update_ratio: 0.05
18
+ train.batch_size: 1
19
+ train.compute_type: bf16
20
+ train.create_new_adapter: false
21
+ train.cutoff_len: 4096
22
+ train.dataset:
23
+ - historical_sleep_story
24
+ train.dataset_dir: data
25
+ train.ds_offload: false
26
+ train.ds_stage: none
27
+ train.enable_thinking: false
28
+ train.extra_args: '{"optim": "adamw_torch", "gradient_checkpointing": true}'
29
+ train.freeze_extra_modules: ''
30
+ train.freeze_language_model: false
31
+ train.freeze_multi_modal_projector: true
32
+ train.freeze_trainable_layers: 2
33
+ train.freeze_trainable_modules: all
34
+ train.freeze_vision_tower: true
35
+ train.galore_rank: 16
36
+ train.galore_scale: 2
37
+ train.galore_target: all
38
+ train.galore_update_interval: 200
39
+ train.gradient_accumulation_steps: 64
40
+ train.image_max_pixels: 768*768
41
+ train.image_min_pixels: 32*32
42
+ train.learning_rate: 5e-5
43
+ train.logging_steps: 5
44
+ train.lora_alpha: 64
45
+ train.lora_dropout: 0.1
46
+ train.lora_rank: 32
47
+ train.lora_target: ''
48
+ train.loraplus_lr_ratio: 0
49
+ train.lr_scheduler_type: cosine
50
+ train.mask_history: false
51
+ train.max_grad_norm: '1.0'
52
+ train.max_samples: '100000'
53
+ train.neat_packing: false
54
+ train.neftune_alpha: 0
55
+ train.num_train_epochs: '20'
56
+ train.packing: false
57
+ train.ppo_score_norm: false
58
+ train.ppo_whiten_rewards: false
59
+ train.pref_beta: 0.1
60
+ train.pref_ftx: 0
61
+ train.pref_loss: sigmoid
62
+ train.report_to: none
63
+ train.resize_vocab: false
64
+ train.reward_model: []
65
+ train.save_steps: 20
66
+ train.swanlab_api_key: ''
67
+ train.swanlab_link: null
68
+ train.swanlab_mode: cloud
69
+ train.swanlab_project: llamafactory
70
+ train.swanlab_run_name: ''
71
+ train.swanlab_workspace: ''
72
+ train.train_on_prompt: false
73
+ train.training_stage: Supervised Fine-Tuning
74
+ train.use_apollo: false
75
+ train.use_badam: false
76
+ train.use_dora: false
77
+ train.use_galore: false
78
+ train.use_llama_pro: false
79
+ train.use_pissa: false
80
+ train.use_rslora: false
81
+ train.use_swanlab: false
82
+ train.val_size: 0
83
+ train.video_max_pixels: 256*256
84
+ train.video_min_pixels: 16*16
85
+ train.warmup_steps: 0
72/running_log.txt ADDED
@@ -0,0 +1,1191 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [INFO|2026-02-13 01:43:22] configuration_utils.py:667 >> loading configuration file config.json from cache at /workspace/.hf_home/hub/models--Qwen--Qwen2.5-72B-Instruct-GPTQ-Int4/snapshots/da6e9d45661b91e02782f4ae2c6bb39c4a5b4821/config.json
2
+ [INFO|2026-02-13 01:43:22] configuration_utils.py:739 >> Model config Qwen2Config {
3
+ "architectures": [
4
+ "Qwen2ForCausalLM"
5
+ ],
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 151643,
8
+ "dtype": "float16",
9
+ "eos_token_id": 151645,
10
+ "hidden_act": "silu",
11
+ "hidden_size": 8192,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 29696,
14
+ "layer_types": [
15
+ "full_attention",
16
+ "full_attention",
17
+ "full_attention",
18
+ "full_attention",
19
+ "full_attention",
20
+ "full_attention",
21
+ "full_attention",
22
+ "full_attention",
23
+ "full_attention",
24
+ "full_attention",
25
+ "full_attention",
26
+ "full_attention",
27
+ "full_attention",
28
+ "full_attention",
29
+ "full_attention",
30
+ "full_attention",
31
+ "full_attention",
32
+ "full_attention",
33
+ "full_attention",
34
+ "full_attention",
35
+ "full_attention",
36
+ "full_attention",
37
+ "full_attention",
38
+ "full_attention",
39
+ "full_attention",
40
+ "full_attention",
41
+ "full_attention",
42
+ "full_attention",
43
+ "full_attention",
44
+ "full_attention",
45
+ "full_attention",
46
+ "full_attention",
47
+ "full_attention",
48
+ "full_attention",
49
+ "full_attention",
50
+ "full_attention",
51
+ "full_attention",
52
+ "full_attention",
53
+ "full_attention",
54
+ "full_attention",
55
+ "full_attention",
56
+ "full_attention",
57
+ "full_attention",
58
+ "full_attention",
59
+ "full_attention",
60
+ "full_attention",
61
+ "full_attention",
62
+ "full_attention",
63
+ "full_attention",
64
+ "full_attention",
65
+ "full_attention",
66
+ "full_attention",
67
+ "full_attention",
68
+ "full_attention",
69
+ "full_attention",
70
+ "full_attention",
71
+ "full_attention",
72
+ "full_attention",
73
+ "full_attention",
74
+ "full_attention",
75
+ "full_attention",
76
+ "full_attention",
77
+ "full_attention",
78
+ "full_attention",
79
+ "full_attention",
80
+ "full_attention",
81
+ "full_attention",
82
+ "full_attention",
83
+ "full_attention",
84
+ "full_attention",
85
+ "full_attention",
86
+ "full_attention",
87
+ "full_attention",
88
+ "full_attention",
89
+ "full_attention",
90
+ "full_attention",
91
+ "full_attention",
92
+ "full_attention",
93
+ "full_attention",
94
+ "full_attention"
95
+ ],
96
+ "max_position_embeddings": 32768,
97
+ "max_window_layers": 70,
98
+ "model_type": "qwen2",
99
+ "num_attention_heads": 64,
100
+ "num_hidden_layers": 80,
101
+ "num_key_value_heads": 8,
102
+ "pad_token_id": null,
103
+ "quantization_config": {
104
+ "batch_size": 1,
105
+ "bits": 4,
106
+ "block_name_to_quantize": null,
107
+ "cache_block_outputs": true,
108
+ "damp_percent": 0.01,
109
+ "dataset": null,
110
+ "desc_act": false,
111
+ "exllama_config": {
112
+ "version": 1
113
+ },
114
+ "group_size": 128,
115
+ "max_input_length": null,
116
+ "model_seqlen": null,
117
+ "module_name_preceding_first_block": null,
118
+ "modules_in_block_to_quantize": null,
119
+ "pad_token_id": null,
120
+ "quant_method": "gptq",
121
+ "sym": true,
122
+ "tokenizer": null,
123
+ "true_sequential": true,
124
+ "use_cuda_fp16": false,
125
+ "use_exllama": true
126
+ },
127
+ "rms_norm_eps": 1e-06,
128
+ "rope_parameters": {
129
+ "rope_theta": 1000000.0,
130
+ "rope_type": "default"
131
+ },
132
+ "sliding_window": null,
133
+ "tie_word_embeddings": false,
134
+ "transformers_version": "5.0.0",
135
+ "use_cache": true,
136
+ "use_sliding_window": false,
137
+ "vocab_size": 152064
138
+ }
139
+
140
+ [INFO|2026-02-13 01:43:24] configuration_utils.py:667 >> loading configuration file config.json from cache at /workspace/.hf_home/hub/models--Qwen--Qwen2.5-72B-Instruct-GPTQ-Int4/snapshots/da6e9d45661b91e02782f4ae2c6bb39c4a5b4821/config.json
141
+ [INFO|2026-02-13 01:43:24] configuration_utils.py:739 >> Model config Qwen2Config {
142
+ "architectures": [
143
+ "Qwen2ForCausalLM"
144
+ ],
145
+ "attention_dropout": 0.0,
146
+ "bos_token_id": 151643,
147
+ "dtype": "float16",
148
+ "eos_token_id": 151645,
149
+ "hidden_act": "silu",
150
+ "hidden_size": 8192,
151
+ "initializer_range": 0.02,
152
+ "intermediate_size": 29696,
153
+ "layer_types": [
154
+ "full_attention",
155
+ "full_attention",
156
+ "full_attention",
157
+ "full_attention",
158
+ "full_attention",
159
+ "full_attention",
160
+ "full_attention",
161
+ "full_attention",
162
+ "full_attention",
163
+ "full_attention",
164
+ "full_attention",
165
+ "full_attention",
166
+ "full_attention",
167
+ "full_attention",
168
+ "full_attention",
169
+ "full_attention",
170
+ "full_attention",
171
+ "full_attention",
172
+ "full_attention",
173
+ "full_attention",
174
+ "full_attention",
175
+ "full_attention",
176
+ "full_attention",
177
+ "full_attention",
178
+ "full_attention",
179
+ "full_attention",
180
+ "full_attention",
181
+ "full_attention",
182
+ "full_attention",
183
+ "full_attention",
184
+ "full_attention",
185
+ "full_attention",
186
+ "full_attention",
187
+ "full_attention",
188
+ "full_attention",
189
+ "full_attention",
190
+ "full_attention",
191
+ "full_attention",
192
+ "full_attention",
193
+ "full_attention",
194
+ "full_attention",
195
+ "full_attention",
196
+ "full_attention",
197
+ "full_attention",
198
+ "full_attention",
199
+ "full_attention",
200
+ "full_attention",
201
+ "full_attention",
202
+ "full_attention",
203
+ "full_attention",
204
+ "full_attention",
205
+ "full_attention",
206
+ "full_attention",
207
+ "full_attention",
208
+ "full_attention",
209
+ "full_attention",
210
+ "full_attention",
211
+ "full_attention",
212
+ "full_attention",
213
+ "full_attention",
214
+ "full_attention",
215
+ "full_attention",
216
+ "full_attention",
217
+ "full_attention",
218
+ "full_attention",
219
+ "full_attention",
220
+ "full_attention",
221
+ "full_attention",
222
+ "full_attention",
223
+ "full_attention",
224
+ "full_attention",
225
+ "full_attention",
226
+ "full_attention",
227
+ "full_attention",
228
+ "full_attention",
229
+ "full_attention",
230
+ "full_attention",
231
+ "full_attention",
232
+ "full_attention",
233
+ "full_attention"
234
+ ],
235
+ "max_position_embeddings": 32768,
236
+ "max_window_layers": 70,
237
+ "model_type": "qwen2",
238
+ "num_attention_heads": 64,
239
+ "num_hidden_layers": 80,
240
+ "num_key_value_heads": 8,
241
+ "pad_token_id": null,
242
+ "quantization_config": {
243
+ "batch_size": 1,
244
+ "bits": 4,
245
+ "block_name_to_quantize": null,
246
+ "cache_block_outputs": true,
247
+ "damp_percent": 0.01,
248
+ "dataset": null,
249
+ "desc_act": false,
250
+ "exllama_config": {
251
+ "version": 1
252
+ },
253
+ "group_size": 128,
254
+ "max_input_length": null,
255
+ "model_seqlen": null,
256
+ "module_name_preceding_first_block": null,
257
+ "modules_in_block_to_quantize": null,
258
+ "pad_token_id": null,
259
+ "quant_method": "gptq",
260
+ "sym": true,
261
+ "tokenizer": null,
262
+ "true_sequential": true,
263
+ "use_cuda_fp16": false,
264
+ "use_exllama": true
265
+ },
266
+ "rms_norm_eps": 1e-06,
267
+ "rope_parameters": {
268
+ "rope_theta": 1000000.0,
269
+ "rope_type": "default"
270
+ },
271
+ "sliding_window": null,
272
+ "tie_word_embeddings": false,
273
+ "transformers_version": "5.0.0",
274
+ "use_cache": true,
275
+ "use_sliding_window": false,
276
+ "vocab_size": 152064
277
+ }
278
+
279
+ [INFO|2026-02-13 01:43:24] configuration_utils.py:667 >> loading configuration file config.json from cache at /workspace/.hf_home/hub/models--Qwen--Qwen2.5-72B-Instruct-GPTQ-Int4/snapshots/da6e9d45661b91e02782f4ae2c6bb39c4a5b4821/config.json
280
+ [INFO|2026-02-13 01:43:24] configuration_utils.py:739 >> Model config Qwen2Config {
281
+ "architectures": [
282
+ "Qwen2ForCausalLM"
283
+ ],
284
+ "attention_dropout": 0.0,
285
+ "bos_token_id": 151643,
286
+ "dtype": "float16",
287
+ "eos_token_id": 151645,
288
+ "hidden_act": "silu",
289
+ "hidden_size": 8192,
290
+ "initializer_range": 0.02,
291
+ "intermediate_size": 29696,
292
+ "layer_types": [
293
+ "full_attention",
294
+ "full_attention",
295
+ "full_attention",
296
+ "full_attention",
297
+ "full_attention",
298
+ "full_attention",
299
+ "full_attention",
300
+ "full_attention",
301
+ "full_attention",
302
+ "full_attention",
303
+ "full_attention",
304
+ "full_attention",
305
+ "full_attention",
306
+ "full_attention",
307
+ "full_attention",
308
+ "full_attention",
309
+ "full_attention",
310
+ "full_attention",
311
+ "full_attention",
312
+ "full_attention",
313
+ "full_attention",
314
+ "full_attention",
315
+ "full_attention",
316
+ "full_attention",
317
+ "full_attention",
318
+ "full_attention",
319
+ "full_attention",
320
+ "full_attention",
321
+ "full_attention",
322
+ "full_attention",
323
+ "full_attention",
324
+ "full_attention",
325
+ "full_attention",
326
+ "full_attention",
327
+ "full_attention",
328
+ "full_attention",
329
+ "full_attention",
330
+ "full_attention",
331
+ "full_attention",
332
+ "full_attention",
333
+ "full_attention",
334
+ "full_attention",
335
+ "full_attention",
336
+ "full_attention",
337
+ "full_attention",
338
+ "full_attention",
339
+ "full_attention",
340
+ "full_attention",
341
+ "full_attention",
342
+ "full_attention",
343
+ "full_attention",
344
+ "full_attention",
345
+ "full_attention",
346
+ "full_attention",
347
+ "full_attention",
348
+ "full_attention",
349
+ "full_attention",
350
+ "full_attention",
351
+ "full_attention",
352
+ "full_attention",
353
+ "full_attention",
354
+ "full_attention",
355
+ "full_attention",
356
+ "full_attention",
357
+ "full_attention",
358
+ "full_attention",
359
+ "full_attention",
360
+ "full_attention",
361
+ "full_attention",
362
+ "full_attention",
363
+ "full_attention",
364
+ "full_attention",
365
+ "full_attention",
366
+ "full_attention",
367
+ "full_attention",
368
+ "full_attention",
369
+ "full_attention",
370
+ "full_attention",
371
+ "full_attention",
372
+ "full_attention"
373
+ ],
374
+ "max_position_embeddings": 32768,
375
+ "max_window_layers": 70,
376
+ "model_type": "qwen2",
377
+ "num_attention_heads": 64,
378
+ "num_hidden_layers": 80,
379
+ "num_key_value_heads": 8,
380
+ "pad_token_id": null,
381
+ "quantization_config": {
382
+ "batch_size": 1,
383
+ "bits": 4,
384
+ "block_name_to_quantize": null,
385
+ "cache_block_outputs": true,
386
+ "damp_percent": 0.01,
387
+ "dataset": null,
388
+ "desc_act": false,
389
+ "exllama_config": {
390
+ "version": 1
391
+ },
392
+ "group_size": 128,
393
+ "max_input_length": null,
394
+ "model_seqlen": null,
395
+ "module_name_preceding_first_block": null,
396
+ "modules_in_block_to_quantize": null,
397
+ "pad_token_id": null,
398
+ "quant_method": "gptq",
399
+ "sym": true,
400
+ "tokenizer": null,
401
+ "true_sequential": true,
402
+ "use_cuda_fp16": false,
403
+ "use_exllama": true
404
+ },
405
+ "rms_norm_eps": 1e-06,
406
+ "rope_parameters": {
407
+ "rope_theta": 1000000.0,
408
+ "rope_type": "default"
409
+ },
410
+ "sliding_window": null,
411
+ "tie_word_embeddings": false,
412
+ "transformers_version": "5.0.0",
413
+ "use_cache": true,
414
+ "use_sliding_window": false,
415
+ "vocab_size": 152064
416
+ }
417
+
418
+ [INFO|2026-02-13 01:43:25] logging.py:144 >> Loading dataset dataset.json...
419
+ [INFO|2026-02-13 01:43:26] configuration_utils.py:667 >> loading configuration file config.json from cache at /workspace/.hf_home/hub/models--Qwen--Qwen2.5-72B-Instruct-GPTQ-Int4/snapshots/da6e9d45661b91e02782f4ae2c6bb39c4a5b4821/config.json
420
+ [INFO|2026-02-13 01:43:26] configuration_utils.py:739 >> Model config Qwen2Config {
421
+ "architectures": [
422
+ "Qwen2ForCausalLM"
423
+ ],
424
+ "attention_dropout": 0.0,
425
+ "bos_token_id": 151643,
426
+ "dtype": "float16",
427
+ "eos_token_id": 151645,
428
+ "hidden_act": "silu",
429
+ "hidden_size": 8192,
430
+ "initializer_range": 0.02,
431
+ "intermediate_size": 29696,
432
+ "layer_types": [
433
+ "full_attention",
434
+ "full_attention",
435
+ "full_attention",
436
+ "full_attention",
437
+ "full_attention",
438
+ "full_attention",
439
+ "full_attention",
440
+ "full_attention",
441
+ "full_attention",
442
+ "full_attention",
443
+ "full_attention",
444
+ "full_attention",
445
+ "full_attention",
446
+ "full_attention",
447
+ "full_attention",
448
+ "full_attention",
449
+ "full_attention",
450
+ "full_attention",
451
+ "full_attention",
452
+ "full_attention",
453
+ "full_attention",
454
+ "full_attention",
455
+ "full_attention",
456
+ "full_attention",
457
+ "full_attention",
458
+ "full_attention",
459
+ "full_attention",
460
+ "full_attention",
461
+ "full_attention",
462
+ "full_attention",
463
+ "full_attention",
464
+ "full_attention",
465
+ "full_attention",
466
+ "full_attention",
467
+ "full_attention",
468
+ "full_attention",
469
+ "full_attention",
470
+ "full_attention",
471
+ "full_attention",
472
+ "full_attention",
473
+ "full_attention",
474
+ "full_attention",
475
+ "full_attention",
476
+ "full_attention",
477
+ "full_attention",
478
+ "full_attention",
479
+ "full_attention",
480
+ "full_attention",
481
+ "full_attention",
482
+ "full_attention",
483
+ "full_attention",
484
+ "full_attention",
485
+ "full_attention",
486
+ "full_attention",
487
+ "full_attention",
488
+ "full_attention",
489
+ "full_attention",
490
+ "full_attention",
491
+ "full_attention",
492
+ "full_attention",
493
+ "full_attention",
494
+ "full_attention",
495
+ "full_attention",
496
+ "full_attention",
497
+ "full_attention",
498
+ "full_attention",
499
+ "full_attention",
500
+ "full_attention",
501
+ "full_attention",
502
+ "full_attention",
503
+ "full_attention",
504
+ "full_attention",
505
+ "full_attention",
506
+ "full_attention",
507
+ "full_attention",
508
+ "full_attention",
509
+ "full_attention",
510
+ "full_attention",
511
+ "full_attention",
512
+ "full_attention"
513
+ ],
514
+ "max_position_embeddings": 32768,
515
+ "max_window_layers": 70,
516
+ "model_type": "qwen2",
517
+ "num_attention_heads": 64,
518
+ "num_hidden_layers": 80,
519
+ "num_key_value_heads": 8,
520
+ "pad_token_id": null,
521
+ "quantization_config": {
522
+ "batch_size": 1,
523
+ "bits": 4,
524
+ "block_name_to_quantize": null,
525
+ "cache_block_outputs": true,
526
+ "damp_percent": 0.01,
527
+ "dataset": null,
528
+ "desc_act": false,
529
+ "exllama_config": {
530
+ "version": 1
531
+ },
532
+ "group_size": 128,
533
+ "max_input_length": null,
534
+ "model_seqlen": null,
535
+ "module_name_preceding_first_block": null,
536
+ "modules_in_block_to_quantize": null,
537
+ "pad_token_id": null,
538
+ "quant_method": "gptq",
539
+ "sym": true,
540
+ "tokenizer": null,
541
+ "true_sequential": true,
542
+ "use_cuda_fp16": false,
543
+ "use_exllama": true
544
+ },
545
+ "rms_norm_eps": 1e-06,
546
+ "rope_parameters": {
547
+ "rope_theta": 1000000.0,
548
+ "rope_type": "default"
549
+ },
550
+ "sliding_window": null,
551
+ "tie_word_embeddings": false,
552
+ "transformers_version": "5.0.0",
553
+ "use_cache": true,
554
+ "use_sliding_window": false,
555
+ "vocab_size": 152064
556
+ }
557
+
558
+ [INFO|2026-02-13 01:43:26] logging.py:144 >> Loading 4-bit GPTQ-quantized model.
559
+ [INFO|2026-02-13 01:43:26] logging.py:144 >> KV cache is disabled during training.
560
+ [INFO|2026-02-13 01:43:26] auto.py:249 >>
561
+ [INFO|2026-02-13 01:43:27] modeling_utils.py:732 >> loading weights file model.safetensors from cache at /workspace/.hf_home/hub/models--Qwen--Qwen2.5-72B-Instruct-GPTQ-Int4/snapshots/da6e9d45661b91e02782f4ae2c6bb39c4a5b4821/model.safetensors.index.json
562
+ [INFO|2026-02-13 01:45:49] modeling_utils.py:801 >> Will use dtype=torch.float16 as defined in model's config object
563
+ [INFO|2026-02-13 01:45:49] configuration_utils.py:1014 >> Generate config GenerationConfig {
564
+ "bos_token_id": 151643,
565
+ "eos_token_id": 151645,
566
+ "output_attentions": false,
567
+ "output_hidden_states": false,
568
+ "use_cache": false
569
+ }
570
+
571
+ [WARNING|2026-02-13 01:45:50] logging.py:327 >> `loss_type=None` was set in the config but it is unrecognized. Using the default loss: `ForCausalLMLoss`.
572
+ [INFO|2026-02-13 01:45:55] configuration_utils.py:967 >> loading configuration file generation_config.json from cache at /workspace/.hf_home/hub/models--Qwen--Qwen2.5-72B-Instruct-GPTQ-Int4/snapshots/da6e9d45661b91e02782f4ae2c6bb39c4a5b4821/generation_config.json
573
+ [INFO|2026-02-13 01:45:55] configuration_utils.py:1014 >> Generate config GenerationConfig {
574
+ "bos_token_id": 151643,
575
+ "do_sample": true,
576
+ "eos_token_id": [
577
+ 151645,
578
+ 151643
579
+ ],
580
+ "pad_token_id": 151643,
581
+ "repetition_penalty": 1.05,
582
+ "temperature": 0.7,
583
+ "top_k": 20,
584
+ "top_p": 0.8
585
+ }
586
+
587
+ [INFO|2026-02-13 01:45:55] dynamic_module_utils.py:406 >> Could not locate the custom_generate/generate.py inside Qwen/Qwen2.5-72B-Instruct-GPTQ-Int4.
588
+ [INFO|2026-02-13 01:45:56] logging.py:144 >> Gradient checkpointing enabled.
589
+ [INFO|2026-02-13 01:45:56] logging.py:144 >> Using torch SDPA for faster training and inference.
590
+ [INFO|2026-02-13 01:45:56] logging.py:144 >> Upcasting trainable params to float32.
591
+ [INFO|2026-02-13 01:45:56] logging.py:144 >> Fine-tuning method: LoRA
592
+ [INFO|2026-02-13 01:45:56] logging.py:144 >> Found linear modules: q_proj,v_proj,k_proj,gate_proj,down_proj,up_proj,o_proj
593
+ [INFO|2026-02-13 01:45:58] logging.py:144 >> trainable params: 422,051,840 || all params: 2,914,787,328 || trainable%: 14.4797
594
+ [WARNING|2026-02-13 01:45:58] trainer.py:922 >> The tokenizer has new PAD/BOS/EOS tokens that differ from the model config and generation config. The model config and generation config were aligned accordingly, being updated with the tokenizer's values. Updated tokens: {'bos_token_id': None, 'pad_token_id': 151643}.
595
+ [INFO|2026-02-13 01:45:58] trainer.py:2383 >> ***** Running training *****
596
+ [INFO|2026-02-13 01:45:58] trainer.py:2384 >> Num examples = 224
597
+ [INFO|2026-02-13 01:45:58] trainer.py:2385 >> Num Epochs = 20
598
+ [INFO|2026-02-13 01:45:58] trainer.py:2386 >> Instantaneous batch size per device = 1
599
+ [INFO|2026-02-13 01:45:58] trainer.py:2389 >> Total train batch size (w. parallel, distributed & accumulation) = 64
600
+ [INFO|2026-02-13 01:45:58] trainer.py:2390 >> Gradient Accumulation steps = 64
601
+ [INFO|2026-02-13 01:45:58] trainer.py:2391 >> Total optimization steps = 80
602
+ [INFO|2026-02-13 01:45:58] trainer.py:2392 >> Number of trainable parameters = 422,051,840
603
+ [INFO|2026-02-13 02:03:40] logging.py:144 >> {'loss': 2.5576, 'learning_rate': 4.9692e-05, 'epoch': 1.29, 'throughput': 326.14}
604
+ [INFO|2026-02-13 02:21:54] logging.py:144 >> {'loss': 2.4329, 'learning_rate': 4.8455e-05, 'epoch': 2.57, 'throughput': 326.93}
605
+ [INFO|2026-02-13 02:39:14] logging.py:144 >> {'loss': 2.3525, 'learning_rate': 4.6316e-05, 'epoch': 3.86, 'throughput': 325.77}
606
+ [INFO|2026-02-13 02:55:24] logging.py:144 >> {'loss': 2.2965, 'learning_rate': 4.3358e-05, 'epoch': 5.00, 'throughput': 326.17}
607
+ [INFO|2026-02-13 02:55:24] trainer.py:4115 >> Saving model checkpoint to saves/Qwen2.5-72B-Instruct-GPTQ-Int4/lora/72/checkpoint-20
608
+ [INFO|2026-02-13 02:55:24] configuration_utils.py:667 >> loading configuration file config.json from cache at /workspace/.hf_home/hub/models--Qwen--Qwen2.5-72B-Instruct-GPTQ-Int4/snapshots/da6e9d45661b91e02782f4ae2c6bb39c4a5b4821/config.json
609
+ [INFO|2026-02-13 02:55:24] configuration_utils.py:739 >> Model config Qwen2Config {
610
+ "architectures": [
611
+ "Qwen2ForCausalLM"
612
+ ],
613
+ "attention_dropout": 0.0,
614
+ "bos_token_id": 151643,
615
+ "dtype": "float16",
616
+ "eos_token_id": 151645,
617
+ "hidden_act": "silu",
618
+ "hidden_size": 8192,
619
+ "initializer_range": 0.02,
620
+ "intermediate_size": 29696,
621
+ "layer_types": [
622
+ "full_attention",
623
+ "full_attention",
624
+ "full_attention",
625
+ "full_attention",
626
+ "full_attention",
627
+ "full_attention",
628
+ "full_attention",
629
+ "full_attention",
630
+ "full_attention",
631
+ "full_attention",
632
+ "full_attention",
633
+ "full_attention",
634
+ "full_attention",
635
+ "full_attention",
636
+ "full_attention",
637
+ "full_attention",
638
+ "full_attention",
639
+ "full_attention",
640
+ "full_attention",
641
+ "full_attention",
642
+ "full_attention",
643
+ "full_attention",
644
+ "full_attention",
645
+ "full_attention",
646
+ "full_attention",
647
+ "full_attention",
648
+ "full_attention",
649
+ "full_attention",
650
+ "full_attention",
651
+ "full_attention",
652
+ "full_attention",
653
+ "full_attention",
654
+ "full_attention",
655
+ "full_attention",
656
+ "full_attention",
657
+ "full_attention",
658
+ "full_attention",
659
+ "full_attention",
660
+ "full_attention",
661
+ "full_attention",
662
+ "full_attention",
663
+ "full_attention",
664
+ "full_attention",
665
+ "full_attention",
666
+ "full_attention",
667
+ "full_attention",
668
+ "full_attention",
669
+ "full_attention",
670
+ "full_attention",
671
+ "full_attention",
672
+ "full_attention",
673
+ "full_attention",
674
+ "full_attention",
675
+ "full_attention",
676
+ "full_attention",
677
+ "full_attention",
678
+ "full_attention",
679
+ "full_attention",
680
+ "full_attention",
681
+ "full_attention",
682
+ "full_attention",
683
+ "full_attention",
684
+ "full_attention",
685
+ "full_attention",
686
+ "full_attention",
687
+ "full_attention",
688
+ "full_attention",
689
+ "full_attention",
690
+ "full_attention",
691
+ "full_attention",
692
+ "full_attention",
693
+ "full_attention",
694
+ "full_attention",
695
+ "full_attention",
696
+ "full_attention",
697
+ "full_attention",
698
+ "full_attention",
699
+ "full_attention",
700
+ "full_attention",
701
+ "full_attention"
702
+ ],
703
+ "max_position_embeddings": 32768,
704
+ "max_window_layers": 70,
705
+ "model_type": "qwen2",
706
+ "num_attention_heads": 64,
707
+ "num_hidden_layers": 80,
708
+ "num_key_value_heads": 8,
709
+ "pad_token_id": null,
710
+ "quantization_config": {
711
+ "batch_size": 1,
712
+ "bits": 4,
713
+ "block_name_to_quantize": null,
714
+ "cache_block_outputs": true,
715
+ "damp_percent": 0.01,
716
+ "dataset": null,
717
+ "desc_act": false,
718
+ "exllama_config": {
719
+ "version": 1
720
+ },
721
+ "group_size": 128,
722
+ "max_input_length": null,
723
+ "model_seqlen": null,
724
+ "module_name_preceding_first_block": null,
725
+ "modules_in_block_to_quantize": null,
726
+ "pad_token_id": null,
727
+ "quant_method": "gptq",
728
+ "sym": true,
729
+ "tokenizer": null,
730
+ "true_sequential": true,
731
+ "use_cuda_fp16": false,
732
+ "use_exllama": true
733
+ },
734
+ "rms_norm_eps": 1e-06,
735
+ "rope_parameters": {
736
+ "rope_theta": 1000000.0,
737
+ "rope_type": "default"
738
+ },
739
+ "sliding_window": null,
740
+ "tie_word_embeddings": false,
741
+ "transformers_version": "5.0.0",
742
+ "use_cache": true,
743
+ "use_sliding_window": false,
744
+ "vocab_size": 152064
745
+ }
746
+
747
+ [INFO|2026-02-13 02:55:25] tokenization_utils_base.py:3327 >> chat template saved in saves/Qwen2.5-72B-Instruct-GPTQ-Int4/lora/72/checkpoint-20/chat_template.jinja
748
+ [INFO|2026-02-13 02:55:25] tokenization_utils_base.py:2181 >> tokenizer config file saved in saves/Qwen2.5-72B-Instruct-GPTQ-Int4/lora/72/checkpoint-20/tokenizer_config.json
749
+ [INFO|2026-02-13 03:13:07] logging.py:144 >> {'loss': 2.2461, 'learning_rate': 3.9695e-05, 'epoch': 6.29, 'throughput': 325.81}
750
+ [INFO|2026-02-13 03:30:58] logging.py:144 >> {'loss': 2.1893, 'learning_rate': 3.5466e-05, 'epoch': 7.57, 'throughput': 325.78}
751
+ [INFO|2026-02-13 03:49:14] logging.py:144 >> {'loss': 2.1421, 'learning_rate': 3.0836e-05, 'epoch': 8.86, 'throughput': 326.18}
752
+ [INFO|2026-02-13 04:04:54] logging.py:144 >> {'loss': 2.0971, 'learning_rate': 2.5981e-05, 'epoch': 10.00, 'throughput': 326.00}
753
+ [INFO|2026-02-13 04:04:54] trainer.py:4115 >> Saving model checkpoint to saves/Qwen2.5-72B-Instruct-GPTQ-Int4/lora/72/checkpoint-40
754
+ [INFO|2026-02-13 04:04:54] configuration_utils.py:667 >> loading configuration file config.json from cache at /workspace/.hf_home/hub/models--Qwen--Qwen2.5-72B-Instruct-GPTQ-Int4/snapshots/da6e9d45661b91e02782f4ae2c6bb39c4a5b4821/config.json
755
+ [INFO|2026-02-13 04:04:54] configuration_utils.py:739 >> Model config Qwen2Config {
756
+ "architectures": [
757
+ "Qwen2ForCausalLM"
758
+ ],
759
+ "attention_dropout": 0.0,
760
+ "bos_token_id": 151643,
761
+ "dtype": "float16",
762
+ "eos_token_id": 151645,
763
+ "hidden_act": "silu",
764
+ "hidden_size": 8192,
765
+ "initializer_range": 0.02,
766
+ "intermediate_size": 29696,
767
+ "layer_types": [
768
+ "full_attention",
769
+ "full_attention",
770
+ "full_attention",
771
+ "full_attention",
772
+ "full_attention",
773
+ "full_attention",
774
+ "full_attention",
775
+ "full_attention",
776
+ "full_attention",
777
+ "full_attention",
778
+ "full_attention",
779
+ "full_attention",
780
+ "full_attention",
781
+ "full_attention",
782
+ "full_attention",
783
+ "full_attention",
784
+ "full_attention",
785
+ "full_attention",
786
+ "full_attention",
787
+ "full_attention",
788
+ "full_attention",
789
+ "full_attention",
790
+ "full_attention",
791
+ "full_attention",
792
+ "full_attention",
793
+ "full_attention",
794
+ "full_attention",
795
+ "full_attention",
796
+ "full_attention",
797
+ "full_attention",
798
+ "full_attention",
799
+ "full_attention",
800
+ "full_attention",
801
+ "full_attention",
802
+ "full_attention",
803
+ "full_attention",
804
+ "full_attention",
805
+ "full_attention",
806
+ "full_attention",
807
+ "full_attention",
808
+ "full_attention",
809
+ "full_attention",
810
+ "full_attention",
811
+ "full_attention",
812
+ "full_attention",
813
+ "full_attention",
814
+ "full_attention",
815
+ "full_attention",
816
+ "full_attention",
817
+ "full_attention",
818
+ "full_attention",
819
+ "full_attention",
820
+ "full_attention",
821
+ "full_attention",
822
+ "full_attention",
823
+ "full_attention",
824
+ "full_attention",
825
+ "full_attention",
826
+ "full_attention",
827
+ "full_attention",
828
+ "full_attention",
829
+ "full_attention",
830
+ "full_attention",
831
+ "full_attention",
832
+ "full_attention",
833
+ "full_attention",
834
+ "full_attention",
835
+ "full_attention",
836
+ "full_attention",
837
+ "full_attention",
838
+ "full_attention",
839
+ "full_attention",
840
+ "full_attention",
841
+ "full_attention",
842
+ "full_attention",
843
+ "full_attention",
844
+ "full_attention",
845
+ "full_attention",
846
+ "full_attention",
847
+ "full_attention"
848
+ ],
849
+ "max_position_embeddings": 32768,
850
+ "max_window_layers": 70,
851
+ "model_type": "qwen2",
852
+ "num_attention_heads": 64,
853
+ "num_hidden_layers": 80,
854
+ "num_key_value_heads": 8,
855
+ "pad_token_id": null,
856
+ "quantization_config": {
857
+ "batch_size": 1,
858
+ "bits": 4,
859
+ "block_name_to_quantize": null,
860
+ "cache_block_outputs": true,
861
+ "damp_percent": 0.01,
862
+ "dataset": null,
863
+ "desc_act": false,
864
+ "exllama_config": {
865
+ "version": 1
866
+ },
867
+ "group_size": 128,
868
+ "max_input_length": null,
869
+ "model_seqlen": null,
870
+ "module_name_preceding_first_block": null,
871
+ "modules_in_block_to_quantize": null,
872
+ "pad_token_id": null,
873
+ "quant_method": "gptq",
874
+ "sym": true,
875
+ "tokenizer": null,
876
+ "true_sequential": true,
877
+ "use_cuda_fp16": false,
878
+ "use_exllama": true
879
+ },
880
+ "rms_norm_eps": 1e-06,
881
+ "rope_parameters": {
882
+ "rope_theta": 1000000.0,
883
+ "rope_type": "default"
884
+ },
885
+ "sliding_window": null,
886
+ "tie_word_embeddings": false,
887
+ "transformers_version": "5.0.0",
888
+ "use_cache": true,
889
+ "use_sliding_window": false,
890
+ "vocab_size": 152064
891
+ }
892
+
893
+ [INFO|2026-02-13 04:04:55] tokenization_utils_base.py:3327 >> chat template saved in saves/Qwen2.5-72B-Instruct-GPTQ-Int4/lora/72/checkpoint-40/chat_template.jinja
894
+ [INFO|2026-02-13 04:04:55] tokenization_utils_base.py:2181 >> tokenizer config file saved in saves/Qwen2.5-72B-Instruct-GPTQ-Int4/lora/72/checkpoint-40/tokenizer_config.json
895
+ [INFO|2026-02-13 04:22:44] logging.py:144 >> {'loss': 2.0464, 'learning_rate': 2.1089e-05, 'epoch': 11.29, 'throughput': 325.81}
896
+ [INFO|2026-02-13 04:40:59] logging.py:144 >> {'loss': 2.0127, 'learning_rate': 1.6347e-05, 'epoch': 12.57, 'throughput': 325.84}
897
+ [INFO|2026-02-13 04:58:39] logging.py:144 >> {'loss': 1.9826, 'learning_rate': 1.1938e-05, 'epoch': 13.86, 'throughput': 325.78}
898
+ [INFO|2026-02-13 05:14:31] logging.py:144 >> {'loss': 1.9407, 'learning_rate': 8.0300e-06, 'epoch': 15.00, 'throughput': 325.77}
899
+ [INFO|2026-02-13 05:14:31] trainer.py:4115 >> Saving model checkpoint to saves/Qwen2.5-72B-Instruct-GPTQ-Int4/lora/72/checkpoint-60
900
+ [INFO|2026-02-13 05:14:31] configuration_utils.py:667 >> loading configuration file config.json from cache at /workspace/.hf_home/hub/models--Qwen--Qwen2.5-72B-Instruct-GPTQ-Int4/snapshots/da6e9d45661b91e02782f4ae2c6bb39c4a5b4821/config.json
901
+ [INFO|2026-02-13 05:14:31] configuration_utils.py:739 >> Model config Qwen2Config {
902
+ "architectures": [
903
+ "Qwen2ForCausalLM"
904
+ ],
905
+ "attention_dropout": 0.0,
906
+ "bos_token_id": 151643,
907
+ "dtype": "float16",
908
+ "eos_token_id": 151645,
909
+ "hidden_act": "silu",
910
+ "hidden_size": 8192,
911
+ "initializer_range": 0.02,
912
+ "intermediate_size": 29696,
913
+ "layer_types": [
914
+ "full_attention",
915
+ "full_attention",
916
+ "full_attention",
917
+ "full_attention",
918
+ "full_attention",
919
+ "full_attention",
920
+ "full_attention",
921
+ "full_attention",
922
+ "full_attention",
923
+ "full_attention",
924
+ "full_attention",
925
+ "full_attention",
926
+ "full_attention",
927
+ "full_attention",
928
+ "full_attention",
929
+ "full_attention",
930
+ "full_attention",
931
+ "full_attention",
932
+ "full_attention",
933
+ "full_attention",
934
+ "full_attention",
935
+ "full_attention",
936
+ "full_attention",
937
+ "full_attention",
938
+ "full_attention",
939
+ "full_attention",
940
+ "full_attention",
941
+ "full_attention",
942
+ "full_attention",
943
+ "full_attention",
944
+ "full_attention",
945
+ "full_attention",
946
+ "full_attention",
947
+ "full_attention",
948
+ "full_attention",
949
+ "full_attention",
950
+ "full_attention",
951
+ "full_attention",
952
+ "full_attention",
953
+ "full_attention",
954
+ "full_attention",
955
+ "full_attention",
956
+ "full_attention",
957
+ "full_attention",
958
+ "full_attention",
959
+ "full_attention",
960
+ "full_attention",
961
+ "full_attention",
962
+ "full_attention",
963
+ "full_attention",
964
+ "full_attention",
965
+ "full_attention",
966
+ "full_attention",
967
+ "full_attention",
968
+ "full_attention",
969
+ "full_attention",
970
+ "full_attention",
971
+ "full_attention",
972
+ "full_attention",
973
+ "full_attention",
974
+ "full_attention",
975
+ "full_attention",
976
+ "full_attention",
977
+ "full_attention",
978
+ "full_attention",
979
+ "full_attention",
980
+ "full_attention",
981
+ "full_attention",
982
+ "full_attention",
983
+ "full_attention",
984
+ "full_attention",
985
+ "full_attention",
986
+ "full_attention",
987
+ "full_attention",
988
+ "full_attention",
989
+ "full_attention",
990
+ "full_attention",
991
+ "full_attention",
992
+ "full_attention",
993
+ "full_attention"
994
+ ],
995
+ "max_position_embeddings": 32768,
996
+ "max_window_layers": 70,
997
+ "model_type": "qwen2",
998
+ "num_attention_heads": 64,
999
+ "num_hidden_layers": 80,
1000
+ "num_key_value_heads": 8,
1001
+ "pad_token_id": null,
1002
+ "quantization_config": {
1003
+ "batch_size": 1,
1004
+ "bits": 4,
1005
+ "block_name_to_quantize": null,
1006
+ "cache_block_outputs": true,
1007
+ "damp_percent": 0.01,
1008
+ "dataset": null,
1009
+ "desc_act": false,
1010
+ "exllama_config": {
1011
+ "version": 1
1012
+ },
1013
+ "group_size": 128,
1014
+ "max_input_length": null,
1015
+ "model_seqlen": null,
1016
+ "module_name_preceding_first_block": null,
1017
+ "modules_in_block_to_quantize": null,
1018
+ "pad_token_id": null,
1019
+ "quant_method": "gptq",
1020
+ "sym": true,
1021
+ "tokenizer": null,
1022
+ "true_sequential": true,
1023
+ "use_cuda_fp16": false,
1024
+ "use_exllama": true
1025
+ },
1026
+ "rms_norm_eps": 1e-06,
1027
+ "rope_parameters": {
1028
+ "rope_theta": 1000000.0,
1029
+ "rope_type": "default"
1030
+ },
1031
+ "sliding_window": null,
1032
+ "tie_word_embeddings": false,
1033
+ "transformers_version": "5.0.0",
1034
+ "use_cache": true,
1035
+ "use_sliding_window": false,
1036
+ "vocab_size": 152064
1037
+ }
1038
+
1039
+ [INFO|2026-02-13 05:14:32] tokenization_utils_base.py:3327 >> chat template saved in saves/Qwen2.5-72B-Instruct-GPTQ-Int4/lora/72/checkpoint-60/chat_template.jinja
1040
+ [INFO|2026-02-13 05:14:32] tokenization_utils_base.py:2181 >> tokenizer config file saved in saves/Qwen2.5-72B-Instruct-GPTQ-Int4/lora/72/checkpoint-60/tokenizer_config.json
1041
+ [INFO|2026-02-13 05:14:58] trainer.py:2657 >>
1042
+
1043
+ Training completed. Do not forget to share your model on huggingface.co/models =)
1044
+
1045
+
1046
+ [INFO|2026-02-13 05:14:58] trainer.py:4115 >> Saving model checkpoint to saves/Qwen2.5-72B-Instruct-GPTQ-Int4/lora/72
1047
+ [INFO|2026-02-13 05:14:59] configuration_utils.py:667 >> loading configuration file config.json from cache at /workspace/.hf_home/hub/models--Qwen--Qwen2.5-72B-Instruct-GPTQ-Int4/snapshots/da6e9d45661b91e02782f4ae2c6bb39c4a5b4821/config.json
1048
+ [INFO|2026-02-13 05:14:59] configuration_utils.py:739 >> Model config Qwen2Config {
1049
+ "architectures": [
1050
+ "Qwen2ForCausalLM"
1051
+ ],
1052
+ "attention_dropout": 0.0,
1053
+ "bos_token_id": 151643,
1054
+ "dtype": "float16",
1055
+ "eos_token_id": 151645,
1056
+ "hidden_act": "silu",
1057
+ "hidden_size": 8192,
1058
+ "initializer_range": 0.02,
1059
+ "intermediate_size": 29696,
1060
+ "layer_types": [
1061
+ "full_attention",
1062
+ "full_attention",
1063
+ "full_attention",
1064
+ "full_attention",
1065
+ "full_attention",
1066
+ "full_attention",
1067
+ "full_attention",
1068
+ "full_attention",
1069
+ "full_attention",
1070
+ "full_attention",
1071
+ "full_attention",
1072
+ "full_attention",
1073
+ "full_attention",
1074
+ "full_attention",
1075
+ "full_attention",
1076
+ "full_attention",
1077
+ "full_attention",
1078
+ "full_attention",
1079
+ "full_attention",
1080
+ "full_attention",
1081
+ "full_attention",
1082
+ "full_attention",
1083
+ "full_attention",
1084
+ "full_attention",
1085
+ "full_attention",
1086
+ "full_attention",
1087
+ "full_attention",
1088
+ "full_attention",
1089
+ "full_attention",
1090
+ "full_attention",
1091
+ "full_attention",
1092
+ "full_attention",
1093
+ "full_attention",
1094
+ "full_attention",
1095
+ "full_attention",
1096
+ "full_attention",
1097
+ "full_attention",
1098
+ "full_attention",
1099
+ "full_attention",
1100
+ "full_attention",
1101
+ "full_attention",
1102
+ "full_attention",
1103
+ "full_attention",
1104
+ "full_attention",
1105
+ "full_attention",
1106
+ "full_attention",
1107
+ "full_attention",
1108
+ "full_attention",
1109
+ "full_attention",
1110
+ "full_attention",
1111
+ "full_attention",
1112
+ "full_attention",
1113
+ "full_attention",
1114
+ "full_attention",
1115
+ "full_attention",
1116
+ "full_attention",
1117
+ "full_attention",
1118
+ "full_attention",
1119
+ "full_attention",
1120
+ "full_attention",
1121
+ "full_attention",
1122
+ "full_attention",
1123
+ "full_attention",
1124
+ "full_attention",
1125
+ "full_attention",
1126
+ "full_attention",
1127
+ "full_attention",
1128
+ "full_attention",
1129
+ "full_attention",
1130
+ "full_attention",
1131
+ "full_attention",
1132
+ "full_attention",
1133
+ "full_attention",
1134
+ "full_attention",
1135
+ "full_attention",
1136
+ "full_attention",
1137
+ "full_attention",
1138
+ "full_attention",
1139
+ "full_attention",
1140
+ "full_attention"
1141
+ ],
1142
+ "max_position_embeddings": 32768,
1143
+ "max_window_layers": 70,
1144
+ "model_type": "qwen2",
1145
+ "num_attention_heads": 64,
1146
+ "num_hidden_layers": 80,
1147
+ "num_key_value_heads": 8,
1148
+ "pad_token_id": null,
1149
+ "quantization_config": {
1150
+ "batch_size": 1,
1151
+ "bits": 4,
1152
+ "block_name_to_quantize": null,
1153
+ "cache_block_outputs": true,
1154
+ "damp_percent": 0.01,
1155
+ "dataset": null,
1156
+ "desc_act": false,
1157
+ "exllama_config": {
1158
+ "version": 1
1159
+ },
1160
+ "group_size": 128,
1161
+ "max_input_length": null,
1162
+ "model_seqlen": null,
1163
+ "module_name_preceding_first_block": null,
1164
+ "modules_in_block_to_quantize": null,
1165
+ "pad_token_id": null,
1166
+ "quant_method": "gptq",
1167
+ "sym": true,
1168
+ "tokenizer": null,
1169
+ "true_sequential": true,
1170
+ "use_cuda_fp16": false,
1171
+ "use_exllama": true
1172
+ },
1173
+ "rms_norm_eps": 1e-06,
1174
+ "rope_parameters": {
1175
+ "rope_theta": 1000000.0,
1176
+ "rope_type": "default"
1177
+ },
1178
+ "sliding_window": null,
1179
+ "tie_word_embeddings": false,
1180
+ "transformers_version": "5.0.0",
1181
+ "use_cache": true,
1182
+ "use_sliding_window": false,
1183
+ "vocab_size": 152064
1184
+ }
1185
+
1186
+ [INFO|2026-02-13 05:15:00] tokenization_utils_base.py:3327 >> chat template saved in saves/Qwen2.5-72B-Instruct-GPTQ-Int4/lora/72/chat_template.jinja
1187
+ [INFO|2026-02-13 05:15:00] tokenization_utils_base.py:2181 >> tokenizer config file saved in saves/Qwen2.5-72B-Instruct-GPTQ-Int4/lora/72/tokenizer_config.json
1188
+ [WARNING|2026-02-13 05:15:00] logging.py:149 >> No metric eval_loss to plot.
1189
+ [WARNING|2026-02-13 05:15:00] logging.py:149 >> No metric eval_accuracy to plot.
1190
+ [INFO|2026-02-13 05:15:00] modelcard.py:266 >> Dropping the following result as it does not have all the necessary fields:
1191
+ {'task': {'name': 'Causal Language Modeling', 'type': 'text-generation'}}
72/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3fd169731d2cbde95e10bf356d66d5997fd885dd8dbb6fb4684da3f23b2585d8
3
+ size 11421892
72/tokenizer_config.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "backend": "tokenizers",
4
+ "bos_token": null,
5
+ "clean_up_tokenization_spaces": false,
6
+ "eos_token": "<|im_end|>",
7
+ "errors": "replace",
8
+ "extra_special_tokens": [
9
+ "<|im_start|>",
10
+ "<|im_end|>",
11
+ "<|object_ref_start|>",
12
+ "<|object_ref_end|>",
13
+ "<|box_start|>",
14
+ "<|box_end|>",
15
+ "<|quad_start|>",
16
+ "<|quad_end|>",
17
+ "<|vision_start|>",
18
+ "<|vision_end|>",
19
+ "<|vision_pad|>",
20
+ "<|image_pad|>",
21
+ "<|video_pad|>"
22
+ ],
23
+ "is_local": false,
24
+ "model_max_length": 131072,
25
+ "pad_token": "<|endoftext|>",
26
+ "padding_side": "right",
27
+ "split_special_tokens": false,
28
+ "tokenizer_class": "Qwen2Tokenizer",
29
+ "unk_token": null
30
+ }
72/train_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 15.0,
3
+ "num_input_tokens_seen": 4085560,
4
+ "total_flos": 4.09147353759744e+16,
5
+ "train_loss": 2.195078119387229,
6
+ "train_runtime": 12540.6077,
7
+ "train_samples_per_second": 0.357,
8
+ "train_steps_per_second": 0.006
9
+ }
72/trainer_log.jsonl ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"current_steps": 5, "total_steps": 80, "loss": 2.557632255554199, "lr": 4.9692208514878444e-05, "epoch": 1.2857142857142856, "percentage": 6.25, "elapsed_time": "0:17:41", "remaining_time": "4:25:28", "throughput": 326.14, "total_tokens": 346336}
2
+ {"current_steps": 10, "total_steps": 80, "loss": 2.432936096191406, "lr": 4.8454783398062106e-05, "epoch": 2.571428571428571, "percentage": 12.5, "elapsed_time": "0:35:56", "remaining_time": "4:11:35", "throughput": 326.93, "total_tokens": 705016}
3
+ {"current_steps": 15, "total_steps": 80, "loss": 2.3525007247924803, "lr": 4.6316004108852305e-05, "epoch": 3.857142857142857, "percentage": 18.75, "elapsed_time": "0:53:16", "remaining_time": "3:50:51", "throughput": 325.77, "total_tokens": 1041296}
4
+ {"current_steps": 20, "total_steps": 80, "loss": 2.2965370178222657, "lr": 4.335806273589214e-05, "epoch": 5.0, "percentage": 25.0, "elapsed_time": "1:09:25", "remaining_time": "3:28:17", "throughput": 326.17, "total_tokens": 1358800}
5
+ {"current_steps": 25, "total_steps": 80, "loss": 2.2461246490478515, "lr": 3.969463130731183e-05, "epoch": 6.285714285714286, "percentage": 31.25, "elapsed_time": "1:27:09", "remaining_time": "3:11:45", "throughput": 325.81, "total_tokens": 1703832}
6
+ {"current_steps": 30, "total_steps": 80, "loss": 2.1892810821533204, "lr": 3.54664934384357e-05, "epoch": 7.571428571428571, "percentage": 37.5, "elapsed_time": "1:45:00", "remaining_time": "2:55:00", "throughput": 325.78, "total_tokens": 2052472}
7
+ {"current_steps": 35, "total_steps": 80, "loss": 2.1421274185180663, "lr": 3.083613409639764e-05, "epoch": 8.857142857142858, "percentage": 43.75, "elapsed_time": "2:03:15", "remaining_time": "2:38:29", "throughput": 326.18, "total_tokens": 2412408}
8
+ {"current_steps": 40, "total_steps": 80, "loss": 2.097066307067871, "lr": 2.598149539397672e-05, "epoch": 10.0, "percentage": 50.0, "elapsed_time": "2:18:56", "remaining_time": "2:18:56", "throughput": 326.0, "total_tokens": 2717600}
9
+ {"current_steps": 45, "total_steps": 80, "loss": 2.046438980102539, "lr": 2.1089138373994223e-05, "epoch": 11.285714285714286, "percentage": 56.25, "elapsed_time": "2:36:45", "remaining_time": "2:01:55", "throughput": 325.81, "total_tokens": 3064544}
10
+ {"current_steps": 50, "total_steps": 80, "loss": 2.0126678466796877, "lr": 1.6347073573062672e-05, "epoch": 12.571428571428571, "percentage": 62.5, "elapsed_time": "2:55:00", "remaining_time": "1:45:00", "throughput": 325.84, "total_tokens": 3421576}
11
+ {"current_steps": 55, "total_steps": 80, "loss": 1.9826169967651368, "lr": 1.1937535882101281e-05, "epoch": 13.857142857142858, "percentage": 68.75, "elapsed_time": "3:12:40", "remaining_time": "1:27:34", "throughput": 325.78, "total_tokens": 3766296}
12
+ {"current_steps": 60, "total_steps": 80, "loss": 1.9407329559326172, "lr": 8.029981361676456e-06, "epoch": 15.0, "percentage": 75.0, "elapsed_time": "3:28:33", "remaining_time": "1:09:31", "throughput": 325.77, "total_tokens": 4076400}
13
+ {"current_steps": 60, "total_steps": 80, "epoch": 15.0, "percentage": 75.0, "elapsed_time": "3:29:00", "remaining_time": "1:09:40", "throughput": 325.79, "total_tokens": 4085560}
72/trainer_state.json ADDED
@@ -0,0 +1,164 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 15.0,
6
+ "eval_steps": 500,
7
+ "global_step": 60,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 1.2857142857142856,
14
+ "grad_norm": 0.2836259603500366,
15
+ "learning_rate": 4.9692208514878444e-05,
16
+ "loss": 2.557632255554199,
17
+ "num_input_tokens_seen": 346336,
18
+ "step": 5,
19
+ "train_runtime": 1061.9262,
20
+ "train_tokens_per_second": 326.139
21
+ },
22
+ {
23
+ "epoch": 2.571428571428571,
24
+ "grad_norm": 0.20870207250118256,
25
+ "learning_rate": 4.8454783398062106e-05,
26
+ "loss": 2.432936096191406,
27
+ "num_input_tokens_seen": 705016,
28
+ "step": 10,
29
+ "train_runtime": 2156.4566,
30
+ "train_tokens_per_second": 326.933
31
+ },
32
+ {
33
+ "epoch": 3.857142857142857,
34
+ "grad_norm": 0.1494155377149582,
35
+ "learning_rate": 4.6316004108852305e-05,
36
+ "loss": 2.3525007247924803,
37
+ "num_input_tokens_seen": 1041296,
38
+ "step": 15,
39
+ "train_runtime": 3196.4466,
40
+ "train_tokens_per_second": 325.767
41
+ },
42
+ {
43
+ "epoch": 5.0,
44
+ "grad_norm": 0.17206469178199768,
45
+ "learning_rate": 4.335806273589214e-05,
46
+ "loss": 2.2965370178222657,
47
+ "num_input_tokens_seen": 1358800,
48
+ "step": 20,
49
+ "train_runtime": 4165.9384,
50
+ "train_tokens_per_second": 326.169
51
+ },
52
+ {
53
+ "epoch": 6.285714285714286,
54
+ "grad_norm": 0.14012837409973145,
55
+ "learning_rate": 3.969463130731183e-05,
56
+ "loss": 2.2461246490478515,
57
+ "num_input_tokens_seen": 1703832,
58
+ "step": 25,
59
+ "train_runtime": 5229.5525,
60
+ "train_tokens_per_second": 325.808
61
+ },
62
+ {
63
+ "epoch": 7.571428571428571,
64
+ "grad_norm": 0.15643367171287537,
65
+ "learning_rate": 3.54664934384357e-05,
66
+ "loss": 2.1892810821533204,
67
+ "num_input_tokens_seen": 2052472,
68
+ "step": 30,
69
+ "train_runtime": 6300.2335,
70
+ "train_tokens_per_second": 325.777
71
+ },
72
+ {
73
+ "epoch": 8.857142857142858,
74
+ "grad_norm": 0.1400991678237915,
75
+ "learning_rate": 3.083613409639764e-05,
76
+ "loss": 2.1421274185180663,
77
+ "num_input_tokens_seen": 2412408,
78
+ "step": 35,
79
+ "train_runtime": 7395.9727,
80
+ "train_tokens_per_second": 326.179
81
+ },
82
+ {
83
+ "epoch": 10.0,
84
+ "grad_norm": 0.24552510678768158,
85
+ "learning_rate": 2.598149539397672e-05,
86
+ "loss": 2.097066307067871,
87
+ "num_input_tokens_seen": 2717600,
88
+ "step": 40,
89
+ "train_runtime": 8336.2619,
90
+ "train_tokens_per_second": 325.997
91
+ },
92
+ {
93
+ "epoch": 11.285714285714286,
94
+ "grad_norm": 0.17399123311042786,
95
+ "learning_rate": 2.1089138373994223e-05,
96
+ "loss": 2.046438980102539,
97
+ "num_input_tokens_seen": 3064544,
98
+ "step": 45,
99
+ "train_runtime": 9405.8661,
100
+ "train_tokens_per_second": 325.812
101
+ },
102
+ {
103
+ "epoch": 12.571428571428571,
104
+ "grad_norm": 0.19238384068012238,
105
+ "learning_rate": 1.6347073573062672e-05,
106
+ "loss": 2.0126678466796877,
107
+ "num_input_tokens_seen": 3421576,
108
+ "step": 50,
109
+ "train_runtime": 10500.9313,
110
+ "train_tokens_per_second": 325.835
111
+ },
112
+ {
113
+ "epoch": 13.857142857142858,
114
+ "grad_norm": 0.2132018357515335,
115
+ "learning_rate": 1.1937535882101281e-05,
116
+ "loss": 1.9826169967651368,
117
+ "num_input_tokens_seen": 3766296,
118
+ "step": 55,
119
+ "train_runtime": 11560.8898,
120
+ "train_tokens_per_second": 325.779
121
+ },
122
+ {
123
+ "epoch": 15.0,
124
+ "grad_norm": 0.3054792881011963,
125
+ "learning_rate": 8.029981361676456e-06,
126
+ "loss": 1.9407329559326172,
127
+ "num_input_tokens_seen": 4076400,
128
+ "step": 60,
129
+ "train_runtime": 12513.0566,
130
+ "train_tokens_per_second": 325.772
131
+ },
132
+ {
133
+ "epoch": 15.0,
134
+ "num_input_tokens_seen": 4085560,
135
+ "step": 60,
136
+ "total_flos": 4.09147353759744e+16,
137
+ "train_loss": 2.195078119387229,
138
+ "train_runtime": 12540.6077,
139
+ "train_samples_per_second": 0.357,
140
+ "train_steps_per_second": 0.006
141
+ }
142
+ ],
143
+ "logging_steps": 5,
144
+ "max_steps": 80,
145
+ "num_input_tokens_seen": 4085560,
146
+ "num_train_epochs": 20,
147
+ "save_steps": 20,
148
+ "stateful_callbacks": {
149
+ "TrainerControl": {
150
+ "args": {
151
+ "should_epoch_stop": false,
152
+ "should_evaluate": false,
153
+ "should_log": false,
154
+ "should_save": true,
155
+ "should_training_stop": false
156
+ },
157
+ "attributes": {}
158
+ }
159
+ },
160
+ "total_flos": 4.09147353759744e+16,
161
+ "train_batch_size": 1,
162
+ "trial_name": null,
163
+ "trial_params": null
164
+ }
72/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b70be08810f61028ef84ab71566ed75f3e27e3ed7cd48a3b96c9124bd7cadbbb
3
+ size 5585
72/training_args.yaml ADDED
@@ -0,0 +1,35 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ bf16: true
2
+ cutoff_len: 4096
3
+ dataset: historical_sleep_story
4
+ dataset_dir: data
5
+ ddp_timeout: 180000000
6
+ do_train: true
7
+ enable_thinking: false
8
+ finetuning_type: lora
9
+ flash_attn: auto
10
+ gradient_accumulation_steps: 64
11
+ gradient_checkpointing: true
12
+ include_num_input_tokens_seen: true
13
+ learning_rate: 5.0e-05
14
+ logging_steps: 5
15
+ lora_alpha: 64
16
+ lora_dropout: 0.1
17
+ lora_rank: 32
18
+ lora_target: all
19
+ lr_scheduler_type: cosine
20
+ max_grad_norm: 1.0
21
+ max_samples: 100000
22
+ model_name_or_path: Qwen/Qwen2.5-72B-Instruct-GPTQ-Int4
23
+ num_train_epochs: 20.0
24
+ optim: adamw_torch
25
+ output_dir: saves/Qwen2.5-72B-Instruct-GPTQ-Int4/lora/72
26
+ packing: false
27
+ per_device_train_batch_size: 1
28
+ plot_loss: true
29
+ preprocessing_num_workers: 16
30
+ report_to: none
31
+ save_steps: 20
32
+ stage: sft
33
+ template: qwen
34
+ trust_remote_code: true
35
+ warmup_steps: 0
72/training_loss.png ADDED

Git LFS Details

  • SHA256: 03d914470c9fd7ba2d10d7e752f61e41af93555ffc32b5bb7e7bb027a53beaf1
  • Pointer size: 130 Bytes
  • Size of remote file: 35.8 kB