TongZheng1999 commited on
Commit
5263d76
·
verified ·
1 Parent(s): c132b51

Model save

Browse files
README.md ADDED
@@ -0,0 +1,57 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: google/gemma-2-9b-it
3
+ library_name: transformers
4
+ model_name: gemma-2-9b-it-star-code-v1_10-5-3Rounds-iter-1
5
+ tags:
6
+ - generated_from_trainer
7
+ - trl
8
+ - sft
9
+ licence: license
10
+ ---
11
+
12
+ # Model Card for gemma-2-9b-it-star-code-v1_10-5-3Rounds-iter-1
13
+
14
+ This model is a fine-tuned version of [google/gemma-2-9b-it](https://huggingface.co/google/gemma-2-9b-it).
15
+ It has been trained using [TRL](https://github.com/huggingface/trl).
16
+
17
+ ## Quick start
18
+
19
+ ```python
20
+ from transformers import pipeline
21
+
22
+ question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
23
+ generator = pipeline("text-generation", model="TongZheng1999/gemma-2-9b-it-star-code-v1_10-5-3Rounds-iter-1", device="cuda")
24
+ output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
25
+ print(output["generated_text"])
26
+ ```
27
+
28
+ ## Training procedure
29
+
30
+ [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/kidzheng/huggingface/runs/99rj3hgy)
31
+
32
+ This model was trained with SFT.
33
+
34
+ ### Framework versions
35
+
36
+ - TRL: 0.12.0
37
+ - Transformers: 4.49.0
38
+ - Pytorch: 2.6.0
39
+ - Datasets: 3.3.2
40
+ - Tokenizers: 0.21.0
41
+
42
+ ## Citations
43
+
44
+
45
+
46
+ Cite TRL as:
47
+
48
+ ```bibtex
49
+ @misc{vonwerra2022trl,
50
+ title = {{TRL: Transformer Reinforcement Learning}},
51
+ author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallouédec},
52
+ year = 2020,
53
+ journal = {GitHub repository},
54
+ publisher = {GitHub},
55
+ howpublished = {\url{https://github.com/huggingface/trl}}
56
+ }
57
+ ```
all_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 4.0,
3
+ "total_flos": 9.994818287147418e+16,
4
+ "train_loss": 0.3279022932052612,
5
+ "train_runtime": 3201.833,
6
+ "train_samples": 534,
7
+ "train_samples_per_second": 0.834,
8
+ "train_steps_per_second": 0.006
9
+ }
generation_config.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 2,
4
+ "cache_implementation": "hybrid",
5
+ "eos_token_id": 1,
6
+ "pad_token_id": 0,
7
+ "transformers_version": "4.49.0"
8
+ }
train_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 4.0,
3
+ "total_flos": 9.994818287147418e+16,
4
+ "train_loss": 0.3279022932052612,
5
+ "train_runtime": 3201.833,
6
+ "train_samples": 534,
7
+ "train_samples_per_second": 0.834,
8
+ "train_steps_per_second": 0.006
9
+ }
trainer_state.json ADDED
@@ -0,0 +1,77 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 4.0,
5
+ "eval_steps": 500,
6
+ "global_step": 20,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.23880597014925373,
13
+ "grad_norm": 29.275244557684626,
14
+ "learning_rate": 4.9692208514878445e-06,
15
+ "loss": 1.492,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 1.0,
20
+ "grad_norm": 4.877218753196121,
21
+ "learning_rate": 4.267766952966369e-06,
22
+ "loss": 0.4997,
23
+ "step": 5
24
+ },
25
+ {
26
+ "epoch": 2.0,
27
+ "grad_norm": 0.7924658588380961,
28
+ "learning_rate": 2.5e-06,
29
+ "loss": 0.2348,
30
+ "step": 10
31
+ },
32
+ {
33
+ "epoch": 3.0,
34
+ "grad_norm": 0.6495439150180484,
35
+ "learning_rate": 7.322330470336314e-07,
36
+ "loss": 0.1975,
37
+ "step": 15
38
+ },
39
+ {
40
+ "epoch": 4.0,
41
+ "grad_norm": 0.5855051339722586,
42
+ "learning_rate": 0.0,
43
+ "loss": 0.1812,
44
+ "step": 20
45
+ },
46
+ {
47
+ "epoch": 4.0,
48
+ "step": 20,
49
+ "total_flos": 9.994818287147418e+16,
50
+ "train_loss": 0.3279022932052612,
51
+ "train_runtime": 3201.833,
52
+ "train_samples_per_second": 0.834,
53
+ "train_steps_per_second": 0.006
54
+ }
55
+ ],
56
+ "logging_steps": 5,
57
+ "max_steps": 20,
58
+ "num_input_tokens_seen": 0,
59
+ "num_train_epochs": 5,
60
+ "save_steps": 10,
61
+ "stateful_callbacks": {
62
+ "TrainerControl": {
63
+ "args": {
64
+ "should_epoch_stop": false,
65
+ "should_evaluate": false,
66
+ "should_log": false,
67
+ "should_save": true,
68
+ "should_training_stop": true
69
+ },
70
+ "attributes": {}
71
+ }
72
+ },
73
+ "total_flos": 9.994818287147418e+16,
74
+ "train_batch_size": 1,
75
+ "trial_name": null,
76
+ "trial_params": null
77
+ }