sedrickkeh commited on
Commit
9d525dd
·
verified ·
1 Parent(s): ee99988

End of training

Browse files
README.md CHANGED
@@ -4,6 +4,7 @@ license: llama3.1
4
  base_model: meta-llama/Meta-Llama-3.1-8B
5
  tags:
6
  - llama-factory
 
7
  - generated_from_trainer
8
  model-index:
9
  - name: stackexchange_physics
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  # stackexchange_physics
17
 
18
- This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B) on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
  - Loss: 0.8525
21
 
 
4
  base_model: meta-llama/Meta-Llama-3.1-8B
5
  tags:
6
  - llama-factory
7
+ - full
8
  - generated_from_trainer
9
  model-index:
10
  - name: stackexchange_physics
 
16
 
17
  # stackexchange_physics
18
 
19
+ This model is a fine-tuned version of [meta-llama/Meta-Llama-3.1-8B](https://huggingface.co/meta-llama/Meta-Llama-3.1-8B) on the mlfoundations-dev/stackexchange_physics dataset.
20
  It achieves the following results on the evaluation set:
21
  - Loss: 0.8525
22
 
all_results.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.979310344827586,
3
+ "eval_loss": 0.8525338172912598,
4
+ "eval_runtime": 25.539,
5
+ "eval_samples_per_second": 38.255,
6
+ "eval_steps_per_second": 0.626,
7
+ "total_flos": 180694642851840.0,
8
+ "train_loss": 0.8558892673916287,
9
+ "train_runtime": 3801.0513,
10
+ "train_samples_per_second": 14.649,
11
+ "train_steps_per_second": 0.028
12
+ }
eval_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.979310344827586,
3
+ "eval_loss": 0.8525338172912598,
4
+ "eval_runtime": 25.539,
5
+ "eval_samples_per_second": 38.255,
6
+ "eval_steps_per_second": 0.626
7
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.979310344827586,
3
+ "total_flos": 180694642851840.0,
4
+ "train_loss": 0.8558892673916287,
5
+ "train_runtime": 3801.0513,
6
+ "train_samples_per_second": 14.649,
7
+ "train_steps_per_second": 0.028
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,136 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.979310344827586,
5
+ "eval_steps": 500,
6
+ "global_step": 108,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.27586206896551724,
13
+ "grad_norm": 7.99272917412018,
14
+ "learning_rate": 5e-06,
15
+ "loss": 1.0372,
16
+ "step": 10
17
+ },
18
+ {
19
+ "epoch": 0.5517241379310345,
20
+ "grad_norm": 1.3943848882036196,
21
+ "learning_rate": 5e-06,
22
+ "loss": 0.9401,
23
+ "step": 20
24
+ },
25
+ {
26
+ "epoch": 0.8275862068965517,
27
+ "grad_norm": 1.0575396228689353,
28
+ "learning_rate": 5e-06,
29
+ "loss": 0.9015,
30
+ "step": 30
31
+ },
32
+ {
33
+ "epoch": 0.993103448275862,
34
+ "eval_loss": 0.8885926604270935,
35
+ "eval_runtime": 26.5641,
36
+ "eval_samples_per_second": 36.779,
37
+ "eval_steps_per_second": 0.602,
38
+ "step": 36
39
+ },
40
+ {
41
+ "epoch": 1.103448275862069,
42
+ "grad_norm": 1.2477575212494822,
43
+ "learning_rate": 5e-06,
44
+ "loss": 0.936,
45
+ "step": 40
46
+ },
47
+ {
48
+ "epoch": 1.3793103448275863,
49
+ "grad_norm": 1.0469251972786897,
50
+ "learning_rate": 5e-06,
51
+ "loss": 0.8346,
52
+ "step": 50
53
+ },
54
+ {
55
+ "epoch": 1.6551724137931034,
56
+ "grad_norm": 0.9988632319097552,
57
+ "learning_rate": 5e-06,
58
+ "loss": 0.8219,
59
+ "step": 60
60
+ },
61
+ {
62
+ "epoch": 1.9310344827586206,
63
+ "grad_norm": 0.9140527353187675,
64
+ "learning_rate": 5e-06,
65
+ "loss": 0.8158,
66
+ "step": 70
67
+ },
68
+ {
69
+ "epoch": 1.986206896551724,
70
+ "eval_loss": 0.8567079305648804,
71
+ "eval_runtime": 25.9826,
72
+ "eval_samples_per_second": 37.602,
73
+ "eval_steps_per_second": 0.616,
74
+ "step": 72
75
+ },
76
+ {
77
+ "epoch": 2.206896551724138,
78
+ "grad_norm": 1.5245621594542598,
79
+ "learning_rate": 5e-06,
80
+ "loss": 0.843,
81
+ "step": 80
82
+ },
83
+ {
84
+ "epoch": 2.4827586206896552,
85
+ "grad_norm": 1.0372960320477767,
86
+ "learning_rate": 5e-06,
87
+ "loss": 0.7585,
88
+ "step": 90
89
+ },
90
+ {
91
+ "epoch": 2.7586206896551726,
92
+ "grad_norm": 0.9305737492177628,
93
+ "learning_rate": 5e-06,
94
+ "loss": 0.7606,
95
+ "step": 100
96
+ },
97
+ {
98
+ "epoch": 2.979310344827586,
99
+ "eval_loss": 0.8525338172912598,
100
+ "eval_runtime": 24.5522,
101
+ "eval_samples_per_second": 39.793,
102
+ "eval_steps_per_second": 0.652,
103
+ "step": 108
104
+ },
105
+ {
106
+ "epoch": 2.979310344827586,
107
+ "step": 108,
108
+ "total_flos": 180694642851840.0,
109
+ "train_loss": 0.8558892673916287,
110
+ "train_runtime": 3801.0513,
111
+ "train_samples_per_second": 14.649,
112
+ "train_steps_per_second": 0.028
113
+ }
114
+ ],
115
+ "logging_steps": 10,
116
+ "max_steps": 108,
117
+ "num_input_tokens_seen": 0,
118
+ "num_train_epochs": 3,
119
+ "save_steps": 500,
120
+ "stateful_callbacks": {
121
+ "TrainerControl": {
122
+ "args": {
123
+ "should_epoch_stop": false,
124
+ "should_evaluate": false,
125
+ "should_log": false,
126
+ "should_save": true,
127
+ "should_training_stop": true
128
+ },
129
+ "attributes": {}
130
+ }
131
+ },
132
+ "total_flos": 180694642851840.0,
133
+ "train_batch_size": 8,
134
+ "trial_name": null,
135
+ "trial_params": null
136
+ }
training_eval_loss.png ADDED
training_loss.png ADDED