scottsuk0306 commited on
Commit
4f20f56
1 Parent(s): b518a8e

Model save

Browse files
README.md ADDED
@@ -0,0 +1,76 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: transformers
3
+ license: apache-2.0
4
+ base_model: alignment-handbook/zephyr-7b-sft-full
5
+ tags:
6
+ - trl
7
+ - sft
8
+ - generated_from_trainer
9
+ datasets:
10
+ - generator
11
+ model-index:
12
+ - name: zephyr-7b-stem-case-3
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
+ # zephyr-7b-stem-case-3
20
+
21
+ This model is a fine-tuned version of [alignment-handbook/zephyr-7b-sft-full](https://huggingface.co/alignment-handbook/zephyr-7b-sft-full) on the generator dataset.
22
+ It achieves the following results on the evaluation set:
23
+ - Loss: 0.0406
24
+
25
+ ## Model description
26
+
27
+ More information needed
28
+
29
+ ## Intended uses & limitations
30
+
31
+ More information needed
32
+
33
+ ## Training and evaluation data
34
+
35
+ More information needed
36
+
37
+ ## Training procedure
38
+
39
+ ### Training hyperparameters
40
+
41
+ The following hyperparameters were used during training:
42
+ - learning_rate: 1e-05
43
+ - train_batch_size: 8
44
+ - eval_batch_size: 8
45
+ - seed: 42
46
+ - distributed_type: multi-GPU
47
+ - num_devices: 4
48
+ - total_train_batch_size: 32
49
+ - total_eval_batch_size: 32
50
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
51
+ - lr_scheduler_type: cosine
52
+ - lr_scheduler_warmup_ratio: 0.03
53
+ - num_epochs: 10
54
+
55
+ ### Training results
56
+
57
+ | Training Loss | Epoch | Step | Validation Loss |
58
+ |:-------------:|:-----:|:----:|:---------------:|
59
+ | 1.1002 | 1.0 | 2 | 1.2712 |
60
+ | 1.1002 | 2.0 | 4 | 0.8801 |
61
+ | 1.0859 | 3.0 | 6 | 0.6101 |
62
+ | 1.0859 | 4.0 | 8 | 0.4419 |
63
+ | 0.5465 | 5.0 | 10 | 0.2620 |
64
+ | 0.5465 | 6.0 | 12 | 0.1413 |
65
+ | 0.5465 | 7.0 | 14 | 0.0808 |
66
+ | 0.1621 | 8.0 | 16 | 0.0535 |
67
+ | 0.1621 | 9.0 | 18 | 0.0429 |
68
+ | 0.0504 | 10.0 | 20 | 0.0406 |
69
+
70
+
71
+ ### Framework versions
72
+
73
+ - Transformers 4.44.2
74
+ - Pytorch 2.4.1+cu121
75
+ - Datasets 3.0.0
76
+ - Tokenizers 0.19.1
all_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 10.0,
3
+ "total_flos": 1.1184101603357491e+17,
4
+ "train_loss": 0.46195218414068223,
5
+ "train_runtime": 339.5674,
6
+ "train_samples": 848,
7
+ "train_samples_per_second": 1.472,
8
+ "train_steps_per_second": 0.059
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": "4.44.2"
6
+ }
train_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 10.0,
3
+ "total_flos": 1.1184101603357491e+17,
4
+ "train_loss": 0.46195218414068223,
5
+ "train_runtime": 339.5674,
6
+ "train_samples": 848,
7
+ "train_samples_per_second": 1.472,
8
+ "train_steps_per_second": 0.059
9
+ }
trainer_state.json ADDED
@@ -0,0 +1,157 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 10.0,
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": 0.5,
13
+ "grad_norm": 18.081775665283203,
14
+ "learning_rate": 1e-05,
15
+ "loss": 1.1002,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 1.0,
20
+ "eval_loss": 1.2711809873580933,
21
+ "eval_runtime": 5.1274,
22
+ "eval_samples_per_second": 9.752,
23
+ "eval_steps_per_second": 0.39,
24
+ "step": 2
25
+ },
26
+ {
27
+ "epoch": 2.0,
28
+ "eval_loss": 0.8800549507141113,
29
+ "eval_runtime": 5.1265,
30
+ "eval_samples_per_second": 9.753,
31
+ "eval_steps_per_second": 0.39,
32
+ "step": 4
33
+ },
34
+ {
35
+ "epoch": 2.5,
36
+ "grad_norm": 19.409046173095703,
37
+ "learning_rate": 8.94570254698197e-06,
38
+ "loss": 1.0859,
39
+ "step": 5
40
+ },
41
+ {
42
+ "epoch": 3.0,
43
+ "eval_loss": 0.6101047992706299,
44
+ "eval_runtime": 5.1466,
45
+ "eval_samples_per_second": 9.715,
46
+ "eval_steps_per_second": 0.389,
47
+ "step": 6
48
+ },
49
+ {
50
+ "epoch": 4.0,
51
+ "eval_loss": 0.44190600514411926,
52
+ "eval_runtime": 5.1566,
53
+ "eval_samples_per_second": 9.696,
54
+ "eval_steps_per_second": 0.388,
55
+ "step": 8
56
+ },
57
+ {
58
+ "epoch": 5.0,
59
+ "grad_norm": 7.709940433502197,
60
+ "learning_rate": 5.412896727361663e-06,
61
+ "loss": 0.5465,
62
+ "step": 10
63
+ },
64
+ {
65
+ "epoch": 5.0,
66
+ "eval_loss": 0.26203182339668274,
67
+ "eval_runtime": 5.1422,
68
+ "eval_samples_per_second": 9.724,
69
+ "eval_steps_per_second": 0.389,
70
+ "step": 10
71
+ },
72
+ {
73
+ "epoch": 6.0,
74
+ "eval_loss": 0.14132770895957947,
75
+ "eval_runtime": 5.1462,
76
+ "eval_samples_per_second": 9.716,
77
+ "eval_steps_per_second": 0.389,
78
+ "step": 12
79
+ },
80
+ {
81
+ "epoch": 7.0,
82
+ "eval_loss": 0.08078073710203171,
83
+ "eval_runtime": 5.1421,
84
+ "eval_samples_per_second": 9.724,
85
+ "eval_steps_per_second": 0.389,
86
+ "step": 14
87
+ },
88
+ {
89
+ "epoch": 7.5,
90
+ "grad_norm": 1.5183640718460083,
91
+ "learning_rate": 1.6135921418712959e-06,
92
+ "loss": 0.1621,
93
+ "step": 15
94
+ },
95
+ {
96
+ "epoch": 8.0,
97
+ "eval_loss": 0.05351748317480087,
98
+ "eval_runtime": 5.1503,
99
+ "eval_samples_per_second": 9.708,
100
+ "eval_steps_per_second": 0.388,
101
+ "step": 16
102
+ },
103
+ {
104
+ "epoch": 9.0,
105
+ "eval_loss": 0.042922407388687134,
106
+ "eval_runtime": 5.1564,
107
+ "eval_samples_per_second": 9.697,
108
+ "eval_steps_per_second": 0.388,
109
+ "step": 18
110
+ },
111
+ {
112
+ "epoch": 10.0,
113
+ "grad_norm": 0.851169228553772,
114
+ "learning_rate": 0.0,
115
+ "loss": 0.0504,
116
+ "step": 20
117
+ },
118
+ {
119
+ "epoch": 10.0,
120
+ "eval_loss": 0.04061064124107361,
121
+ "eval_runtime": 5.054,
122
+ "eval_samples_per_second": 9.893,
123
+ "eval_steps_per_second": 0.396,
124
+ "step": 20
125
+ },
126
+ {
127
+ "epoch": 10.0,
128
+ "step": 20,
129
+ "total_flos": 1.1184101603357491e+17,
130
+ "train_loss": 0.46195218414068223,
131
+ "train_runtime": 339.5674,
132
+ "train_samples_per_second": 1.472,
133
+ "train_steps_per_second": 0.059
134
+ }
135
+ ],
136
+ "logging_steps": 5,
137
+ "max_steps": 20,
138
+ "num_input_tokens_seen": 0,
139
+ "num_train_epochs": 10,
140
+ "save_steps": 100,
141
+ "stateful_callbacks": {
142
+ "TrainerControl": {
143
+ "args": {
144
+ "should_epoch_stop": false,
145
+ "should_evaluate": false,
146
+ "should_log": false,
147
+ "should_save": true,
148
+ "should_training_stop": true
149
+ },
150
+ "attributes": {}
151
+ }
152
+ },
153
+ "total_flos": 1.1184101603357491e+17,
154
+ "train_batch_size": 8,
155
+ "trial_name": null,
156
+ "trial_params": null
157
+ }