Hugging Face's logo

interview-eval
/
zephyr-7b-sft-full

Text Generation
Transformers
Safetensors
mistral
alignment-handbook
trl
sft
Generated from Trainer
conversational
text-generation-inference
Model card Files Community
Juyoung Suk commited on
Commit
7161976
·
verified ·
1 Parent(s): c9f79bb

Model save

Browse files
Files changed (5) hide show
  1. README.md +76 -0
  2. all_results.json +9 -0
  3. generation_config.json +6 -0
  4. train_results.json +9 -0
  5. trainer_state.json +143 -0
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-sft-full
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-sft-full
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: 0.4854
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: 16
44
+ - eval_batch_size: 8
45
+ - seed: 42
46
+ - distributed_type: multi-GPU
47
+ - num_devices: 8
48
+ - total_train_batch_size: 128
49
+ - total_eval_batch_size: 64
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
+ | 1.1139 | 1.0 | 1 | 1.1133 |
60
+ | 1.1139 | 2.0 | 2 | 1.2947 |
61
+ | 1.1139 | 3.0 | 3 | 1.0839 |
62
+ | 1.1139 | 4.0 | 4 | 0.8506 |
63
+ | 1.0867 | 5.0 | 5 | 0.7521 |
64
+ | 1.0867 | 6.0 | 6 | 0.6283 |
65
+ | 1.0867 | 7.0 | 7 | 0.5677 |
66
+ | 1.0867 | 8.0 | 8 | 0.5233 |
67
+ | 1.0867 | 9.0 | 9 | 0.4948 |
68
+ | 0.5927 | 10.0 | 10 | 0.4854 |
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": 8375186227200.0,
4
+ "train_loss": 0.8424163699150086,
5
+ "train_runtime": 304.1051,
6
+ "train_samples": 851,
7
+ "train_samples_per_second": 1.578,
8
+ "train_steps_per_second": 0.033
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": 8375186227200.0,
4
+ "train_loss": 0.8424163699150086,
5
+ "train_runtime": 304.1051,
6
+ "train_samples": 851,
7
+ "train_samples_per_second": 1.578,
8
+ "train_steps_per_second": 0.033
9
+ }
trainer_state.json ADDED
@@ -0,0 +1,143 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 10.0,
5
+ "eval_steps": 500,
6
+ "global_step": 10,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 1.0,
13
+ "grad_norm": 18.364192038860114,
14
+ "learning_rate": 1e-05,
15
+ "loss": 1.1139,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 1.0,
20
+ "eval_loss": 1.1133455038070679,
21
+ "eval_runtime": 2.8092,
22
+ "eval_samples_per_second": 17.087,
23
+ "eval_steps_per_second": 0.356,
24
+ "step": 1
25
+ },
26
+ {
27
+ "epoch": 2.0,
28
+ "eval_loss": 1.2946728467941284,
29
+ "eval_runtime": 2.6987,
30
+ "eval_samples_per_second": 17.786,
31
+ "eval_steps_per_second": 0.371,
32
+ "step": 2
33
+ },
34
+ {
35
+ "epoch": 3.0,
36
+ "eval_loss": 1.0838984251022339,
37
+ "eval_runtime": 2.6695,
38
+ "eval_samples_per_second": 17.981,
39
+ "eval_steps_per_second": 0.375,
40
+ "step": 3
41
+ },
42
+ {
43
+ "epoch": 4.0,
44
+ "eval_loss": 0.8506386280059814,
45
+ "eval_runtime": 2.6924,
46
+ "eval_samples_per_second": 17.828,
47
+ "eval_steps_per_second": 0.371,
48
+ "step": 4
49
+ },
50
+ {
51
+ "epoch": 5.0,
52
+ "grad_norm": 19.328671929154766,
53
+ "learning_rate": 5.8682408883346535e-06,
54
+ "loss": 1.0867,
55
+ "step": 5
56
+ },
57
+ {
58
+ "epoch": 5.0,
59
+ "eval_loss": 0.7521028518676758,
60
+ "eval_runtime": 2.6708,
61
+ "eval_samples_per_second": 17.972,
62
+ "eval_steps_per_second": 0.374,
63
+ "step": 5
64
+ },
65
+ {
66
+ "epoch": 6.0,
67
+ "eval_loss": 0.628281831741333,
68
+ "eval_runtime": 2.7091,
69
+ "eval_samples_per_second": 17.718,
70
+ "eval_steps_per_second": 0.369,
71
+ "step": 6
72
+ },
73
+ {
74
+ "epoch": 7.0,
75
+ "eval_loss": 0.5677043795585632,
76
+ "eval_runtime": 2.6968,
77
+ "eval_samples_per_second": 17.799,
78
+ "eval_steps_per_second": 0.371,
79
+ "step": 7
80
+ },
81
+ {
82
+ "epoch": 8.0,
83
+ "eval_loss": 0.5233017802238464,
84
+ "eval_runtime": 2.6981,
85
+ "eval_samples_per_second": 17.79,
86
+ "eval_steps_per_second": 0.371,
87
+ "step": 8
88
+ },
89
+ {
90
+ "epoch": 9.0,
91
+ "eval_loss": 0.4948171079158783,
92
+ "eval_runtime": 2.6682,
93
+ "eval_samples_per_second": 17.99,
94
+ "eval_steps_per_second": 0.375,
95
+ "step": 9
96
+ },
97
+ {
98
+ "epoch": 10.0,
99
+ "grad_norm": 4.900067571807948,
100
+ "learning_rate": 0.0,
101
+ "loss": 0.5927,
102
+ "step": 10
103
+ },
104
+ {
105
+ "epoch": 10.0,
106
+ "eval_loss": 0.48538458347320557,
107
+ "eval_runtime": 2.6954,
108
+ "eval_samples_per_second": 17.808,
109
+ "eval_steps_per_second": 0.371,
110
+ "step": 10
111
+ },
112
+ {
113
+ "epoch": 10.0,
114
+ "step": 10,
115
+ "total_flos": 8375186227200.0,
116
+ "train_loss": 0.8424163699150086,
117
+ "train_runtime": 304.1051,
118
+ "train_samples_per_second": 1.578,
119
+ "train_steps_per_second": 0.033
120
+ }
121
+ ],
122
+ "logging_steps": 5,
123
+ "max_steps": 10,
124
+ "num_input_tokens_seen": 0,
125
+ "num_train_epochs": 10,
126
+ "save_steps": 100,
127
+ "stateful_callbacks": {
128
+ "TrainerControl": {
129
+ "args": {
130
+ "should_epoch_stop": false,
131
+ "should_evaluate": false,
132
+ "should_log": false,
133
+ "should_save": true,
134
+ "should_training_stop": true
135
+ },
136
+ "attributes": {}
137
+ }
138
+ },
139
+ "total_flos": 8375186227200.0,
140
+ "train_batch_size": 16,
141
+ "trial_name": null,
142
+ "trial_params": null
143
+ }