Update model
Browse files
checkpoint-125/trainer_state.json
ADDED
@@ -0,0 +1,75 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 1.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 125,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.16,
|
13 |
+
"grad_norm": 5.664922714233398,
|
14 |
+
"learning_rate": 4.36e-05,
|
15 |
+
"loss": 11.8337,
|
16 |
+
"step": 20
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.32,
|
20 |
+
"grad_norm": 5.513271808624268,
|
21 |
+
"learning_rate": 3.56e-05,
|
22 |
+
"loss": 5.7797,
|
23 |
+
"step": 40
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.48,
|
27 |
+
"grad_norm": 5.666042327880859,
|
28 |
+
"learning_rate": 2.7600000000000003e-05,
|
29 |
+
"loss": 4.4415,
|
30 |
+
"step": 60
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.64,
|
34 |
+
"grad_norm": 4.240084648132324,
|
35 |
+
"learning_rate": 1.9600000000000002e-05,
|
36 |
+
"loss": 3.9991,
|
37 |
+
"step": 80
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.8,
|
41 |
+
"grad_norm": 3.6688475608825684,
|
42 |
+
"learning_rate": 1.16e-05,
|
43 |
+
"loss": 3.8011,
|
44 |
+
"step": 100
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.96,
|
48 |
+
"grad_norm": 4.186811447143555,
|
49 |
+
"learning_rate": 3.6e-06,
|
50 |
+
"loss": 3.6621,
|
51 |
+
"step": 120
|
52 |
+
}
|
53 |
+
],
|
54 |
+
"logging_steps": 20,
|
55 |
+
"max_steps": 125,
|
56 |
+
"num_input_tokens_seen": 0,
|
57 |
+
"num_train_epochs": 1,
|
58 |
+
"save_steps": 60000,
|
59 |
+
"stateful_callbacks": {
|
60 |
+
"TrainerControl": {
|
61 |
+
"args": {
|
62 |
+
"should_epoch_stop": false,
|
63 |
+
"should_evaluate": false,
|
64 |
+
"should_log": false,
|
65 |
+
"should_save": true,
|
66 |
+
"should_training_stop": true
|
67 |
+
},
|
68 |
+
"attributes": {}
|
69 |
+
}
|
70 |
+
},
|
71 |
+
"total_flos": 5314359328768000.0,
|
72 |
+
"train_batch_size": 2,
|
73 |
+
"trial_name": null,
|
74 |
+
"trial_params": null
|
75 |
+
}
|