Hugging Face's logo

AmberYifan
/
mistral-gsm8k-sft-full

Safetensors
mistral
alignment-handbook
trl
sft
Generated from Trainer
Model card Files Community
AmberYifan commited on
Commit
796f74c
•
1 Parent(s): f86a92d

Model save

Browse files
Files changed (5) hide show
  1. README.md +66 -0
  2. all_results.json +9 -0
  3. generation_config.json +6 -0
  4. train_results.json +9 -0
  5. trainer_state.json +92 -0
README.md ADDED
@@ -0,0 +1,66 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ base_model: mistralai/Mistral-7B-v0.1
4
+ tags:
5
+ - trl
6
+ - sft
7
+ - generated_from_trainer
8
+ datasets:
9
+ - generator
10
+ model-index:
11
+ - name: mistral-gsm8k-sft-full
12
+ results: []
13
+ ---
14
+
15
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
16
+ should probably proofread and complete it, then remove this comment. -->
17
+
18
+ # mistral-gsm8k-sft-full
19
+
20
+ This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the generator dataset.
21
+ It achieves the following results on the evaluation set:
22
+ - Loss: 0.7393
23
+
24
+ ## Model description
25
+
26
+ More information needed
27
+
28
+ ## Intended uses & limitations
29
+
30
+ More information needed
31
+
32
+ ## Training and evaluation data
33
+
34
+ More information needed
35
+
36
+ ## Training procedure
37
+
38
+ ### Training hyperparameters
39
+
40
+ The following hyperparameters were used during training:
41
+ - learning_rate: 2e-05
42
+ - train_batch_size: 8
43
+ - eval_batch_size: 8
44
+ - seed: 42
45
+ - distributed_type: multi-GPU
46
+ - num_devices: 4
47
+ - total_train_batch_size: 32
48
+ - total_eval_batch_size: 32
49
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
50
+ - lr_scheduler_type: cosine
51
+ - lr_scheduler_warmup_ratio: 0.1
52
+ - num_epochs: 1
53
+
54
+ ### Training results
55
+
56
+ | Training Loss | Epoch | Step | Validation Loss |
57
+ |:-------------:|:-----:|:----:|:---------------:|
58
+ | 0.7412 | 1.0 | 25 | 0.7393 |
59
+
60
+
61
+ ### Framework versions
62
+
63
+ - Transformers 4.43.3
64
+ - Pytorch 2.2.2+cu121
65
+ - Datasets 2.20.0
66
+ - Tokenizers 0.19.1
all_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "total_flos": 2617245696000.0,
4
+ "train_loss": 1.022568130493164,
5
+ "train_runtime": 720.6069,
6
+ "train_samples": 7473,
7
+ "train_samples_per_second": 1.106,
8
+ "train_steps_per_second": 0.035
9
+ }
generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 1,
4
+ "eos_token_id": 2,
5
+ "transformers_version": "4.43.3"
6
+ }
train_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "total_flos": 2617245696000.0,
4
+ "train_loss": 1.022568130493164,
5
+ "train_runtime": 720.6069,
6
+ "train_samples": 7473,
7
+ "train_samples_per_second": 1.106,
8
+ "train_steps_per_second": 0.035
9
+ }
trainer_state.json ADDED
@@ -0,0 +1,92 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.0,
5
+ "eval_steps": 500,
6
+ "global_step": 25,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.04,
13
+ "grad_norm": 52.35670654945737,
14
+ "learning_rate": 6.666666666666667e-06,
15
+ "loss": 1.1797,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.2,
20
+ "grad_norm": 78.00343229930253,
21
+ "learning_rate": 1.9594929736144978e-05,
22
+ "loss": 1.3873,
23
+ "step": 5
24
+ },
25
+ {
26
+ "epoch": 0.4,
27
+ "grad_norm": 14.842921808281147,
28
+ "learning_rate": 1.5406408174555978e-05,
29
+ "loss": 1.2452,
30
+ "step": 10
31
+ },
32
+ {
33
+ "epoch": 0.6,
34
+ "grad_norm": 5.790210129797682,
35
+ "learning_rate": 8.576851617267151e-06,
36
+ "loss": 0.9707,
37
+ "step": 15
38
+ },
39
+ {
40
+ "epoch": 0.8,
41
+ "grad_norm": 3.7307757411742117,
42
+ "learning_rate": 2.4425042564574186e-06,
43
+ "loss": 0.81,
44
+ "step": 20
45
+ },
46
+ {
47
+ "epoch": 1.0,
48
+ "grad_norm": 2.349583644410224,
49
+ "learning_rate": 0.0,
50
+ "loss": 0.7412,
51
+ "step": 25
52
+ },
53
+ {
54
+ "epoch": 1.0,
55
+ "eval_loss": 0.7393369078636169,
56
+ "eval_runtime": 15.2556,
57
+ "eval_samples_per_second": 9.374,
58
+ "eval_steps_per_second": 0.328,
59
+ "step": 25
60
+ },
61
+ {
62
+ "epoch": 1.0,
63
+ "step": 25,
64
+ "total_flos": 2617245696000.0,
65
+ "train_loss": 1.022568130493164,
66
+ "train_runtime": 720.6069,
67
+ "train_samples_per_second": 1.106,
68
+ "train_steps_per_second": 0.035
69
+ }
70
+ ],
71
+ "logging_steps": 5,
72
+ "max_steps": 25,
73
+ "num_input_tokens_seen": 0,
74
+ "num_train_epochs": 1,
75
+ "save_steps": 800,
76
+ "stateful_callbacks": {
77
+ "TrainerControl": {
78
+ "args": {
79
+ "should_epoch_stop": false,
80
+ "should_evaluate": false,
81
+ "should_log": false,
82
+ "should_save": true,
83
+ "should_training_stop": true
84
+ },
85
+ "attributes": {}
86
+ }
87
+ },
88
+ "total_flos": 2617245696000.0,
89
+ "train_batch_size": 8,
90
+ "trial_name": null,
91
+ "trial_params": null
92
+ }