add model files
Browse files- config.json +44 -0
- rng_state.pth +0 -0
- scheduler.pt +0 -0
- trainer_state.json +1416 -0
- training_args.bin +0 -0
config.json
ADDED
@@ -0,0 +1,44 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "aubmindlab/aragpt2-base",
|
3 |
+
"activation_function": "gelu_new",
|
4 |
+
"architectures": [
|
5 |
+
"GPT2LMHeadModel"
|
6 |
+
],
|
7 |
+
"attn_pdrop": 0.1,
|
8 |
+
"bos_token_id": 0,
|
9 |
+
"embd_pdrop": 0.1,
|
10 |
+
"eos_token_id": 0,
|
11 |
+
"gradient_checkpointing": false,
|
12 |
+
"initializer_range": 0.02,
|
13 |
+
"layer_norm_epsilon": 1e-05,
|
14 |
+
"model_type": "gpt2",
|
15 |
+
"n_ctx": 1024,
|
16 |
+
"n_embd": 768,
|
17 |
+
"n_head": 12,
|
18 |
+
"n_inner": null,
|
19 |
+
"n_layer": 12,
|
20 |
+
"n_positions": 1024,
|
21 |
+
"reorder_and_upcast_attn": false,
|
22 |
+
"resid_pdrop": 0.1,
|
23 |
+
"scale_attn_by_inverse_layer_idx": false,
|
24 |
+
"scale_attn_weights": true,
|
25 |
+
"summary_activation": null,
|
26 |
+
"summary_first_dropout": 0.1,
|
27 |
+
"summary_proj_to_labels": true,
|
28 |
+
"summary_type": "cls_index",
|
29 |
+
"summary_use_proj": true,
|
30 |
+
"task_specific_params": {
|
31 |
+
"text-generation": {
|
32 |
+
"do_sample": true,
|
33 |
+
"max_length": 50,
|
34 |
+
"no_repeat_ngram_size": 3,
|
35 |
+
"num_beams": 5,
|
36 |
+
"repetition_penalty": 3.0,
|
37 |
+
"top_p": 0.95
|
38 |
+
}
|
39 |
+
},
|
40 |
+
"torch_dtype": "float32",
|
41 |
+
"transformers_version": "4.16.2",
|
42 |
+
"use_cache": true,
|
43 |
+
"vocab_size": 64000
|
44 |
+
}
|
rng_state.pth
ADDED
Binary file (14.5 kB). View file
|
|
scheduler.pt
ADDED
Binary file (623 Bytes). View file
|
|
trainer_state.json
ADDED
@@ -0,0 +1,1416 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 5.275652862041678,
|
5 |
+
"global_step": 100000,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.05,
|
12 |
+
"learning_rate": 8.333333333333334e-06,
|
13 |
+
"loss": 6.6159,
|
14 |
+
"step": 1000
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.05,
|
18 |
+
"eval_loss": 5.667357444763184,
|
19 |
+
"eval_runtime": 77.1533,
|
20 |
+
"eval_samples_per_second": 53.413,
|
21 |
+
"eval_steps_per_second": 3.344,
|
22 |
+
"step": 1000
|
23 |
+
},
|
24 |
+
{
|
25 |
+
"epoch": 0.11,
|
26 |
+
"learning_rate": 1.6666666666666667e-05,
|
27 |
+
"loss": 3.8828,
|
28 |
+
"step": 2000
|
29 |
+
},
|
30 |
+
{
|
31 |
+
"epoch": 0.11,
|
32 |
+
"eval_loss": 5.533279895782471,
|
33 |
+
"eval_runtime": 76.0378,
|
34 |
+
"eval_samples_per_second": 54.197,
|
35 |
+
"eval_steps_per_second": 3.393,
|
36 |
+
"step": 2000
|
37 |
+
},
|
38 |
+
{
|
39 |
+
"epoch": 0.16,
|
40 |
+
"learning_rate": 2.5e-05,
|
41 |
+
"loss": 3.571,
|
42 |
+
"step": 3000
|
43 |
+
},
|
44 |
+
{
|
45 |
+
"epoch": 0.16,
|
46 |
+
"eval_loss": 5.119894504547119,
|
47 |
+
"eval_runtime": 76.0046,
|
48 |
+
"eval_samples_per_second": 54.22,
|
49 |
+
"eval_steps_per_second": 3.395,
|
50 |
+
"step": 3000
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 0.21,
|
54 |
+
"learning_rate": 3.3333333333333335e-05,
|
55 |
+
"loss": 3.4645,
|
56 |
+
"step": 4000
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 0.21,
|
60 |
+
"eval_loss": 5.038773536682129,
|
61 |
+
"eval_runtime": 75.6611,
|
62 |
+
"eval_samples_per_second": 54.467,
|
63 |
+
"eval_steps_per_second": 3.41,
|
64 |
+
"step": 4000
|
65 |
+
},
|
66 |
+
{
|
67 |
+
"epoch": 0.26,
|
68 |
+
"learning_rate": 4.166666666666667e-05,
|
69 |
+
"loss": 3.3826,
|
70 |
+
"step": 5000
|
71 |
+
},
|
72 |
+
{
|
73 |
+
"epoch": 0.26,
|
74 |
+
"eval_loss": 4.901891708374023,
|
75 |
+
"eval_runtime": 76.8157,
|
76 |
+
"eval_samples_per_second": 53.648,
|
77 |
+
"eval_steps_per_second": 3.359,
|
78 |
+
"step": 5000
|
79 |
+
},
|
80 |
+
{
|
81 |
+
"epoch": 0.32,
|
82 |
+
"learning_rate": 5e-05,
|
83 |
+
"loss": 3.3503,
|
84 |
+
"step": 6000
|
85 |
+
},
|
86 |
+
{
|
87 |
+
"epoch": 0.32,
|
88 |
+
"eval_loss": 5.015623092651367,
|
89 |
+
"eval_runtime": 76.5299,
|
90 |
+
"eval_samples_per_second": 53.848,
|
91 |
+
"eval_steps_per_second": 3.371,
|
92 |
+
"step": 6000
|
93 |
+
},
|
94 |
+
{
|
95 |
+
"epoch": 0.37,
|
96 |
+
"learning_rate": 4.9727594660855354e-05,
|
97 |
+
"loss": 3.3038,
|
98 |
+
"step": 7000
|
99 |
+
},
|
100 |
+
{
|
101 |
+
"epoch": 0.37,
|
102 |
+
"eval_loss": 4.954394340515137,
|
103 |
+
"eval_runtime": 76.8379,
|
104 |
+
"eval_samples_per_second": 53.632,
|
105 |
+
"eval_steps_per_second": 3.358,
|
106 |
+
"step": 7000
|
107 |
+
},
|
108 |
+
{
|
109 |
+
"epoch": 0.42,
|
110 |
+
"learning_rate": 4.9455189321710707e-05,
|
111 |
+
"loss": 3.2411,
|
112 |
+
"step": 8000
|
113 |
+
},
|
114 |
+
{
|
115 |
+
"epoch": 0.42,
|
116 |
+
"eval_loss": 4.890042304992676,
|
117 |
+
"eval_runtime": 76.8834,
|
118 |
+
"eval_samples_per_second": 53.601,
|
119 |
+
"eval_steps_per_second": 3.356,
|
120 |
+
"step": 8000
|
121 |
+
},
|
122 |
+
{
|
123 |
+
"epoch": 0.47,
|
124 |
+
"learning_rate": 4.918278398256606e-05,
|
125 |
+
"loss": 3.1958,
|
126 |
+
"step": 9000
|
127 |
+
},
|
128 |
+
{
|
129 |
+
"epoch": 0.47,
|
130 |
+
"eval_loss": 5.015548229217529,
|
131 |
+
"eval_runtime": 76.9729,
|
132 |
+
"eval_samples_per_second": 53.538,
|
133 |
+
"eval_steps_per_second": 3.352,
|
134 |
+
"step": 9000
|
135 |
+
},
|
136 |
+
{
|
137 |
+
"epoch": 0.53,
|
138 |
+
"learning_rate": 4.891037864342141e-05,
|
139 |
+
"loss": 3.1856,
|
140 |
+
"step": 10000
|
141 |
+
},
|
142 |
+
{
|
143 |
+
"epoch": 0.53,
|
144 |
+
"eval_loss": 4.8613457679748535,
|
145 |
+
"eval_runtime": 76.9767,
|
146 |
+
"eval_samples_per_second": 53.536,
|
147 |
+
"eval_steps_per_second": 3.352,
|
148 |
+
"step": 10000
|
149 |
+
},
|
150 |
+
{
|
151 |
+
"epoch": 0.58,
|
152 |
+
"learning_rate": 4.863797330427677e-05,
|
153 |
+
"loss": 3.1836,
|
154 |
+
"step": 11000
|
155 |
+
},
|
156 |
+
{
|
157 |
+
"epoch": 0.58,
|
158 |
+
"eval_loss": 4.832852840423584,
|
159 |
+
"eval_runtime": 76.9463,
|
160 |
+
"eval_samples_per_second": 53.557,
|
161 |
+
"eval_steps_per_second": 3.353,
|
162 |
+
"step": 11000
|
163 |
+
},
|
164 |
+
{
|
165 |
+
"epoch": 0.63,
|
166 |
+
"learning_rate": 4.836556796513212e-05,
|
167 |
+
"loss": 3.1566,
|
168 |
+
"step": 12000
|
169 |
+
},
|
170 |
+
{
|
171 |
+
"epoch": 0.63,
|
172 |
+
"eval_loss": 4.755970001220703,
|
173 |
+
"eval_runtime": 76.8635,
|
174 |
+
"eval_samples_per_second": 53.615,
|
175 |
+
"eval_steps_per_second": 3.357,
|
176 |
+
"step": 12000
|
177 |
+
},
|
178 |
+
{
|
179 |
+
"epoch": 0.69,
|
180 |
+
"learning_rate": 4.809316262598747e-05,
|
181 |
+
"loss": 3.1046,
|
182 |
+
"step": 13000
|
183 |
+
},
|
184 |
+
{
|
185 |
+
"epoch": 0.69,
|
186 |
+
"eval_loss": 4.7649054527282715,
|
187 |
+
"eval_runtime": 76.531,
|
188 |
+
"eval_samples_per_second": 53.847,
|
189 |
+
"eval_steps_per_second": 3.371,
|
190 |
+
"step": 13000
|
191 |
+
},
|
192 |
+
{
|
193 |
+
"epoch": 0.74,
|
194 |
+
"learning_rate": 4.7820757286842826e-05,
|
195 |
+
"loss": 3.109,
|
196 |
+
"step": 14000
|
197 |
+
},
|
198 |
+
{
|
199 |
+
"epoch": 0.74,
|
200 |
+
"eval_loss": 4.723949909210205,
|
201 |
+
"eval_runtime": 76.9877,
|
202 |
+
"eval_samples_per_second": 53.528,
|
203 |
+
"eval_steps_per_second": 3.351,
|
204 |
+
"step": 14000
|
205 |
+
},
|
206 |
+
{
|
207 |
+
"epoch": 0.79,
|
208 |
+
"learning_rate": 4.754835194769818e-05,
|
209 |
+
"loss": 3.0862,
|
210 |
+
"step": 15000
|
211 |
+
},
|
212 |
+
{
|
213 |
+
"epoch": 0.79,
|
214 |
+
"eval_loss": 4.741213321685791,
|
215 |
+
"eval_runtime": 77.0649,
|
216 |
+
"eval_samples_per_second": 53.474,
|
217 |
+
"eval_steps_per_second": 3.348,
|
218 |
+
"step": 15000
|
219 |
+
},
|
220 |
+
{
|
221 |
+
"epoch": 0.84,
|
222 |
+
"learning_rate": 4.727594660855353e-05,
|
223 |
+
"loss": 3.0713,
|
224 |
+
"step": 16000
|
225 |
+
},
|
226 |
+
{
|
227 |
+
"epoch": 0.84,
|
228 |
+
"eval_loss": 4.756562232971191,
|
229 |
+
"eval_runtime": 72.732,
|
230 |
+
"eval_samples_per_second": 56.66,
|
231 |
+
"eval_steps_per_second": 3.547,
|
232 |
+
"step": 16000
|
233 |
+
},
|
234 |
+
{
|
235 |
+
"epoch": 0.9,
|
236 |
+
"learning_rate": 4.700354126940888e-05,
|
237 |
+
"loss": 3.052,
|
238 |
+
"step": 17000
|
239 |
+
},
|
240 |
+
{
|
241 |
+
"epoch": 0.9,
|
242 |
+
"eval_loss": 4.908087730407715,
|
243 |
+
"eval_runtime": 71.9667,
|
244 |
+
"eval_samples_per_second": 57.263,
|
245 |
+
"eval_steps_per_second": 3.585,
|
246 |
+
"step": 17000
|
247 |
+
},
|
248 |
+
{
|
249 |
+
"epoch": 0.95,
|
250 |
+
"learning_rate": 4.6731135930264234e-05,
|
251 |
+
"loss": 3.0546,
|
252 |
+
"step": 18000
|
253 |
+
},
|
254 |
+
{
|
255 |
+
"epoch": 0.95,
|
256 |
+
"eval_loss": 4.8161492347717285,
|
257 |
+
"eval_runtime": 72.0137,
|
258 |
+
"eval_samples_per_second": 57.225,
|
259 |
+
"eval_steps_per_second": 3.583,
|
260 |
+
"step": 18000
|
261 |
+
},
|
262 |
+
{
|
263 |
+
"epoch": 1.0,
|
264 |
+
"learning_rate": 4.6458730591119586e-05,
|
265 |
+
"loss": 3.0308,
|
266 |
+
"step": 19000
|
267 |
+
},
|
268 |
+
{
|
269 |
+
"epoch": 1.0,
|
270 |
+
"eval_loss": 4.948371410369873,
|
271 |
+
"eval_runtime": 72.0349,
|
272 |
+
"eval_samples_per_second": 57.208,
|
273 |
+
"eval_steps_per_second": 3.582,
|
274 |
+
"step": 19000
|
275 |
+
},
|
276 |
+
{
|
277 |
+
"epoch": 1.06,
|
278 |
+
"learning_rate": 4.6186325251974945e-05,
|
279 |
+
"loss": 2.864,
|
280 |
+
"step": 20000
|
281 |
+
},
|
282 |
+
{
|
283 |
+
"epoch": 1.06,
|
284 |
+
"eval_loss": 4.853775978088379,
|
285 |
+
"eval_runtime": 72.0568,
|
286 |
+
"eval_samples_per_second": 57.191,
|
287 |
+
"eval_steps_per_second": 3.581,
|
288 |
+
"step": 20000
|
289 |
+
},
|
290 |
+
{
|
291 |
+
"epoch": 1.11,
|
292 |
+
"learning_rate": 4.59139199128303e-05,
|
293 |
+
"loss": 2.8339,
|
294 |
+
"step": 21000
|
295 |
+
},
|
296 |
+
{
|
297 |
+
"epoch": 1.11,
|
298 |
+
"eval_loss": 4.888705730438232,
|
299 |
+
"eval_runtime": 72.0694,
|
300 |
+
"eval_samples_per_second": 57.181,
|
301 |
+
"eval_steps_per_second": 3.58,
|
302 |
+
"step": 21000
|
303 |
+
},
|
304 |
+
{
|
305 |
+
"epoch": 1.16,
|
306 |
+
"learning_rate": 4.564151457368564e-05,
|
307 |
+
"loss": 2.8388,
|
308 |
+
"step": 22000
|
309 |
+
},
|
310 |
+
{
|
311 |
+
"epoch": 1.16,
|
312 |
+
"eval_loss": 4.920900344848633,
|
313 |
+
"eval_runtime": 72.0696,
|
314 |
+
"eval_samples_per_second": 57.181,
|
315 |
+
"eval_steps_per_second": 3.58,
|
316 |
+
"step": 22000
|
317 |
+
},
|
318 |
+
{
|
319 |
+
"epoch": 1.21,
|
320 |
+
"learning_rate": 4.5369109234541e-05,
|
321 |
+
"loss": 2.852,
|
322 |
+
"step": 23000
|
323 |
+
},
|
324 |
+
{
|
325 |
+
"epoch": 1.21,
|
326 |
+
"eval_loss": 4.990002632141113,
|
327 |
+
"eval_runtime": 72.0496,
|
328 |
+
"eval_samples_per_second": 57.197,
|
329 |
+
"eval_steps_per_second": 3.581,
|
330 |
+
"step": 23000
|
331 |
+
},
|
332 |
+
{
|
333 |
+
"epoch": 1.27,
|
334 |
+
"learning_rate": 4.509670389539635e-05,
|
335 |
+
"loss": 2.8415,
|
336 |
+
"step": 24000
|
337 |
+
},
|
338 |
+
{
|
339 |
+
"epoch": 1.27,
|
340 |
+
"eval_loss": 4.912019729614258,
|
341 |
+
"eval_runtime": 72.0569,
|
342 |
+
"eval_samples_per_second": 57.191,
|
343 |
+
"eval_steps_per_second": 3.581,
|
344 |
+
"step": 24000
|
345 |
+
},
|
346 |
+
{
|
347 |
+
"epoch": 1.32,
|
348 |
+
"learning_rate": 4.4824298556251705e-05,
|
349 |
+
"loss": 2.8435,
|
350 |
+
"step": 25000
|
351 |
+
},
|
352 |
+
{
|
353 |
+
"epoch": 1.32,
|
354 |
+
"eval_loss": 4.905246257781982,
|
355 |
+
"eval_runtime": 72.0554,
|
356 |
+
"eval_samples_per_second": 57.192,
|
357 |
+
"eval_steps_per_second": 3.581,
|
358 |
+
"step": 25000
|
359 |
+
},
|
360 |
+
{
|
361 |
+
"epoch": 1.37,
|
362 |
+
"learning_rate": 4.455189321710706e-05,
|
363 |
+
"loss": 2.832,
|
364 |
+
"step": 26000
|
365 |
+
},
|
366 |
+
{
|
367 |
+
"epoch": 1.37,
|
368 |
+
"eval_loss": 5.011730194091797,
|
369 |
+
"eval_runtime": 72.047,
|
370 |
+
"eval_samples_per_second": 57.199,
|
371 |
+
"eval_steps_per_second": 3.581,
|
372 |
+
"step": 26000
|
373 |
+
},
|
374 |
+
{
|
375 |
+
"epoch": 1.42,
|
376 |
+
"learning_rate": 4.427948787796241e-05,
|
377 |
+
"loss": 2.8509,
|
378 |
+
"step": 27000
|
379 |
+
},
|
380 |
+
{
|
381 |
+
"epoch": 1.42,
|
382 |
+
"eval_loss": 5.078315734863281,
|
383 |
+
"eval_runtime": 72.0913,
|
384 |
+
"eval_samples_per_second": 57.164,
|
385 |
+
"eval_steps_per_second": 3.579,
|
386 |
+
"step": 27000
|
387 |
+
},
|
388 |
+
{
|
389 |
+
"epoch": 1.48,
|
390 |
+
"learning_rate": 4.400708253881776e-05,
|
391 |
+
"loss": 2.8436,
|
392 |
+
"step": 28000
|
393 |
+
},
|
394 |
+
{
|
395 |
+
"epoch": 1.48,
|
396 |
+
"eval_loss": 4.865673542022705,
|
397 |
+
"eval_runtime": 72.0906,
|
398 |
+
"eval_samples_per_second": 57.164,
|
399 |
+
"eval_steps_per_second": 3.579,
|
400 |
+
"step": 28000
|
401 |
+
},
|
402 |
+
{
|
403 |
+
"epoch": 1.53,
|
404 |
+
"learning_rate": 4.373467719967311e-05,
|
405 |
+
"loss": 2.8558,
|
406 |
+
"step": 29000
|
407 |
+
},
|
408 |
+
{
|
409 |
+
"epoch": 1.53,
|
410 |
+
"eval_loss": 4.83981990814209,
|
411 |
+
"eval_runtime": 72.0594,
|
412 |
+
"eval_samples_per_second": 57.189,
|
413 |
+
"eval_steps_per_second": 3.58,
|
414 |
+
"step": 29000
|
415 |
+
},
|
416 |
+
{
|
417 |
+
"epoch": 1.58,
|
418 |
+
"learning_rate": 4.346227186052847e-05,
|
419 |
+
"loss": 2.8324,
|
420 |
+
"step": 30000
|
421 |
+
},
|
422 |
+
{
|
423 |
+
"epoch": 1.58,
|
424 |
+
"eval_loss": 4.96191930770874,
|
425 |
+
"eval_runtime": 72.0431,
|
426 |
+
"eval_samples_per_second": 57.202,
|
427 |
+
"eval_steps_per_second": 3.581,
|
428 |
+
"step": 30000
|
429 |
+
},
|
430 |
+
{
|
431 |
+
"epoch": 1.64,
|
432 |
+
"learning_rate": 4.318986652138382e-05,
|
433 |
+
"loss": 2.8234,
|
434 |
+
"step": 31000
|
435 |
+
},
|
436 |
+
{
|
437 |
+
"epoch": 1.64,
|
438 |
+
"eval_loss": 4.830244541168213,
|
439 |
+
"eval_runtime": 72.1429,
|
440 |
+
"eval_samples_per_second": 57.123,
|
441 |
+
"eval_steps_per_second": 3.576,
|
442 |
+
"step": 31000
|
443 |
+
},
|
444 |
+
{
|
445 |
+
"epoch": 1.69,
|
446 |
+
"learning_rate": 4.291746118223917e-05,
|
447 |
+
"loss": 2.8155,
|
448 |
+
"step": 32000
|
449 |
+
},
|
450 |
+
{
|
451 |
+
"epoch": 1.69,
|
452 |
+
"eval_loss": 4.835384845733643,
|
453 |
+
"eval_runtime": 72.0429,
|
454 |
+
"eval_samples_per_second": 57.202,
|
455 |
+
"eval_steps_per_second": 3.581,
|
456 |
+
"step": 32000
|
457 |
+
},
|
458 |
+
{
|
459 |
+
"epoch": 1.74,
|
460 |
+
"learning_rate": 4.264505584309453e-05,
|
461 |
+
"loss": 2.8422,
|
462 |
+
"step": 33000
|
463 |
+
},
|
464 |
+
{
|
465 |
+
"epoch": 1.74,
|
466 |
+
"eval_loss": 5.029502868652344,
|
467 |
+
"eval_runtime": 72.0029,
|
468 |
+
"eval_samples_per_second": 57.234,
|
469 |
+
"eval_steps_per_second": 3.583,
|
470 |
+
"step": 33000
|
471 |
+
},
|
472 |
+
{
|
473 |
+
"epoch": 1.79,
|
474 |
+
"learning_rate": 4.237265050394988e-05,
|
475 |
+
"loss": 2.8192,
|
476 |
+
"step": 34000
|
477 |
+
},
|
478 |
+
{
|
479 |
+
"epoch": 1.79,
|
480 |
+
"eval_loss": 5.054713249206543,
|
481 |
+
"eval_runtime": 72.0962,
|
482 |
+
"eval_samples_per_second": 57.16,
|
483 |
+
"eval_steps_per_second": 3.579,
|
484 |
+
"step": 34000
|
485 |
+
},
|
486 |
+
{
|
487 |
+
"epoch": 1.85,
|
488 |
+
"learning_rate": 4.210024516480523e-05,
|
489 |
+
"loss": 2.839,
|
490 |
+
"step": 35000
|
491 |
+
},
|
492 |
+
{
|
493 |
+
"epoch": 1.85,
|
494 |
+
"eval_loss": 5.000535488128662,
|
495 |
+
"eval_runtime": 72.0721,
|
496 |
+
"eval_samples_per_second": 57.179,
|
497 |
+
"eval_steps_per_second": 3.58,
|
498 |
+
"step": 35000
|
499 |
+
},
|
500 |
+
{
|
501 |
+
"epoch": 1.9,
|
502 |
+
"learning_rate": 4.1827839825660584e-05,
|
503 |
+
"loss": 2.803,
|
504 |
+
"step": 36000
|
505 |
+
},
|
506 |
+
{
|
507 |
+
"epoch": 1.9,
|
508 |
+
"eval_loss": 4.954083442687988,
|
509 |
+
"eval_runtime": 72.0788,
|
510 |
+
"eval_samples_per_second": 57.174,
|
511 |
+
"eval_steps_per_second": 3.579,
|
512 |
+
"step": 36000
|
513 |
+
},
|
514 |
+
{
|
515 |
+
"epoch": 1.95,
|
516 |
+
"learning_rate": 4.1555434486515936e-05,
|
517 |
+
"loss": 2.8096,
|
518 |
+
"step": 37000
|
519 |
+
},
|
520 |
+
{
|
521 |
+
"epoch": 1.95,
|
522 |
+
"eval_loss": 5.285138130187988,
|
523 |
+
"eval_runtime": 72.0675,
|
524 |
+
"eval_samples_per_second": 57.182,
|
525 |
+
"eval_steps_per_second": 3.58,
|
526 |
+
"step": 37000
|
527 |
+
},
|
528 |
+
{
|
529 |
+
"epoch": 2.0,
|
530 |
+
"learning_rate": 4.128302914737129e-05,
|
531 |
+
"loss": 2.7969,
|
532 |
+
"step": 38000
|
533 |
+
},
|
534 |
+
{
|
535 |
+
"epoch": 2.0,
|
536 |
+
"eval_loss": 5.253526210784912,
|
537 |
+
"eval_runtime": 72.067,
|
538 |
+
"eval_samples_per_second": 57.183,
|
539 |
+
"eval_steps_per_second": 3.58,
|
540 |
+
"step": 38000
|
541 |
+
},
|
542 |
+
{
|
543 |
+
"epoch": 2.06,
|
544 |
+
"learning_rate": 4.101062380822665e-05,
|
545 |
+
"loss": 2.61,
|
546 |
+
"step": 39000
|
547 |
+
},
|
548 |
+
{
|
549 |
+
"epoch": 2.06,
|
550 |
+
"eval_loss": 5.254916667938232,
|
551 |
+
"eval_runtime": 72.0845,
|
552 |
+
"eval_samples_per_second": 57.169,
|
553 |
+
"eval_steps_per_second": 3.579,
|
554 |
+
"step": 39000
|
555 |
+
},
|
556 |
+
{
|
557 |
+
"epoch": 2.11,
|
558 |
+
"learning_rate": 4.073821846908199e-05,
|
559 |
+
"loss": 2.6062,
|
560 |
+
"step": 40000
|
561 |
+
},
|
562 |
+
{
|
563 |
+
"epoch": 2.11,
|
564 |
+
"eval_loss": 5.293761730194092,
|
565 |
+
"eval_runtime": 72.1279,
|
566 |
+
"eval_samples_per_second": 57.135,
|
567 |
+
"eval_steps_per_second": 3.577,
|
568 |
+
"step": 40000
|
569 |
+
},
|
570 |
+
{
|
571 |
+
"epoch": 2.16,
|
572 |
+
"learning_rate": 4.0465813129937345e-05,
|
573 |
+
"loss": 2.6025,
|
574 |
+
"step": 41000
|
575 |
+
},
|
576 |
+
{
|
577 |
+
"epoch": 2.16,
|
578 |
+
"eval_loss": 5.301900863647461,
|
579 |
+
"eval_runtime": 77.2131,
|
580 |
+
"eval_samples_per_second": 53.372,
|
581 |
+
"eval_steps_per_second": 3.341,
|
582 |
+
"step": 41000
|
583 |
+
},
|
584 |
+
{
|
585 |
+
"epoch": 2.22,
|
586 |
+
"learning_rate": 4.0193407790792704e-05,
|
587 |
+
"loss": 2.5983,
|
588 |
+
"step": 42000
|
589 |
+
},
|
590 |
+
{
|
591 |
+
"epoch": 2.22,
|
592 |
+
"eval_loss": 5.322200775146484,
|
593 |
+
"eval_runtime": 76.5314,
|
594 |
+
"eval_samples_per_second": 53.847,
|
595 |
+
"eval_steps_per_second": 3.371,
|
596 |
+
"step": 42000
|
597 |
+
},
|
598 |
+
{
|
599 |
+
"epoch": 2.27,
|
600 |
+
"learning_rate": 3.9921002451648056e-05,
|
601 |
+
"loss": 2.6204,
|
602 |
+
"step": 43000
|
603 |
+
},
|
604 |
+
{
|
605 |
+
"epoch": 2.27,
|
606 |
+
"eval_loss": 4.960803031921387,
|
607 |
+
"eval_runtime": 76.6466,
|
608 |
+
"eval_samples_per_second": 53.766,
|
609 |
+
"eval_steps_per_second": 3.366,
|
610 |
+
"step": 43000
|
611 |
+
},
|
612 |
+
{
|
613 |
+
"epoch": 2.32,
|
614 |
+
"learning_rate": 3.964859711250341e-05,
|
615 |
+
"loss": 2.6199,
|
616 |
+
"step": 44000
|
617 |
+
},
|
618 |
+
{
|
619 |
+
"epoch": 2.32,
|
620 |
+
"eval_loss": 5.222067832946777,
|
621 |
+
"eval_runtime": 76.1563,
|
622 |
+
"eval_samples_per_second": 54.112,
|
623 |
+
"eval_steps_per_second": 3.388,
|
624 |
+
"step": 44000
|
625 |
+
},
|
626 |
+
{
|
627 |
+
"epoch": 2.37,
|
628 |
+
"learning_rate": 3.937619177335876e-05,
|
629 |
+
"loss": 2.6125,
|
630 |
+
"step": 45000
|
631 |
+
},
|
632 |
+
{
|
633 |
+
"epoch": 2.37,
|
634 |
+
"eval_loss": 5.104248046875,
|
635 |
+
"eval_runtime": 76.1947,
|
636 |
+
"eval_samples_per_second": 54.085,
|
637 |
+
"eval_steps_per_second": 3.386,
|
638 |
+
"step": 45000
|
639 |
+
},
|
640 |
+
{
|
641 |
+
"epoch": 2.43,
|
642 |
+
"learning_rate": 3.910378643421411e-05,
|
643 |
+
"loss": 2.621,
|
644 |
+
"step": 46000
|
645 |
+
},
|
646 |
+
{
|
647 |
+
"epoch": 2.43,
|
648 |
+
"eval_loss": 5.0914106369018555,
|
649 |
+
"eval_runtime": 76.1296,
|
650 |
+
"eval_samples_per_second": 54.131,
|
651 |
+
"eval_steps_per_second": 3.389,
|
652 |
+
"step": 46000
|
653 |
+
},
|
654 |
+
{
|
655 |
+
"epoch": 2.48,
|
656 |
+
"learning_rate": 3.8831381095069464e-05,
|
657 |
+
"loss": 2.6215,
|
658 |
+
"step": 47000
|
659 |
+
},
|
660 |
+
{
|
661 |
+
"epoch": 2.48,
|
662 |
+
"eval_loss": 5.066018104553223,
|
663 |
+
"eval_runtime": 76.0097,
|
664 |
+
"eval_samples_per_second": 54.217,
|
665 |
+
"eval_steps_per_second": 3.394,
|
666 |
+
"step": 47000
|
667 |
+
},
|
668 |
+
{
|
669 |
+
"epoch": 2.53,
|
670 |
+
"learning_rate": 3.855897575592482e-05,
|
671 |
+
"loss": 2.6388,
|
672 |
+
"step": 48000
|
673 |
+
},
|
674 |
+
{
|
675 |
+
"epoch": 2.53,
|
676 |
+
"eval_loss": 5.130795001983643,
|
677 |
+
"eval_runtime": 76.0566,
|
678 |
+
"eval_samples_per_second": 54.183,
|
679 |
+
"eval_steps_per_second": 3.392,
|
680 |
+
"step": 48000
|
681 |
+
},
|
682 |
+
{
|
683 |
+
"epoch": 2.59,
|
684 |
+
"learning_rate": 3.828657041678017e-05,
|
685 |
+
"loss": 2.6328,
|
686 |
+
"step": 49000
|
687 |
+
},
|
688 |
+
{
|
689 |
+
"epoch": 2.59,
|
690 |
+
"eval_loss": 5.16806173324585,
|
691 |
+
"eval_runtime": 76.22,
|
692 |
+
"eval_samples_per_second": 54.067,
|
693 |
+
"eval_steps_per_second": 3.385,
|
694 |
+
"step": 49000
|
695 |
+
},
|
696 |
+
{
|
697 |
+
"epoch": 2.64,
|
698 |
+
"learning_rate": 3.801416507763552e-05,
|
699 |
+
"loss": 2.6462,
|
700 |
+
"step": 50000
|
701 |
+
},
|
702 |
+
{
|
703 |
+
"epoch": 2.64,
|
704 |
+
"eval_loss": 5.09861946105957,
|
705 |
+
"eval_runtime": 81.8993,
|
706 |
+
"eval_samples_per_second": 50.965,
|
707 |
+
"eval_steps_per_second": 3.187,
|
708 |
+
"step": 50000
|
709 |
+
},
|
710 |
+
{
|
711 |
+
"epoch": 2.69,
|
712 |
+
"learning_rate": 3.774175973849088e-05,
|
713 |
+
"loss": 2.6489,
|
714 |
+
"step": 51000
|
715 |
+
},
|
716 |
+
{
|
717 |
+
"epoch": 2.69,
|
718 |
+
"eval_loss": 5.686895847320557,
|
719 |
+
"eval_runtime": 81.88,
|
720 |
+
"eval_samples_per_second": 50.977,
|
721 |
+
"eval_steps_per_second": 3.188,
|
722 |
+
"step": 51000
|
723 |
+
},
|
724 |
+
{
|
725 |
+
"epoch": 2.74,
|
726 |
+
"learning_rate": 3.746935439934623e-05,
|
727 |
+
"loss": 2.6487,
|
728 |
+
"step": 52000
|
729 |
+
},
|
730 |
+
{
|
731 |
+
"epoch": 2.74,
|
732 |
+
"eval_loss": 5.235321998596191,
|
733 |
+
"eval_runtime": 81.927,
|
734 |
+
"eval_samples_per_second": 50.948,
|
735 |
+
"eval_steps_per_second": 3.186,
|
736 |
+
"step": 52000
|
737 |
+
},
|
738 |
+
{
|
739 |
+
"epoch": 2.8,
|
740 |
+
"learning_rate": 3.719694906020158e-05,
|
741 |
+
"loss": 2.6291,
|
742 |
+
"step": 53000
|
743 |
+
},
|
744 |
+
{
|
745 |
+
"epoch": 2.8,
|
746 |
+
"eval_loss": 5.161744594573975,
|
747 |
+
"eval_runtime": 81.9025,
|
748 |
+
"eval_samples_per_second": 50.963,
|
749 |
+
"eval_steps_per_second": 3.187,
|
750 |
+
"step": 53000
|
751 |
+
},
|
752 |
+
{
|
753 |
+
"epoch": 2.85,
|
754 |
+
"learning_rate": 3.6924543721056935e-05,
|
755 |
+
"loss": 2.6168,
|
756 |
+
"step": 54000
|
757 |
+
},
|
758 |
+
{
|
759 |
+
"epoch": 2.85,
|
760 |
+
"eval_loss": 5.017256259918213,
|
761 |
+
"eval_runtime": 81.8942,
|
762 |
+
"eval_samples_per_second": 50.968,
|
763 |
+
"eval_steps_per_second": 3.187,
|
764 |
+
"step": 54000
|
765 |
+
},
|
766 |
+
{
|
767 |
+
"epoch": 2.9,
|
768 |
+
"learning_rate": 3.665213838191229e-05,
|
769 |
+
"loss": 2.6478,
|
770 |
+
"step": 55000
|
771 |
+
},
|
772 |
+
{
|
773 |
+
"epoch": 2.9,
|
774 |
+
"eval_loss": 4.907939434051514,
|
775 |
+
"eval_runtime": 81.992,
|
776 |
+
"eval_samples_per_second": 50.907,
|
777 |
+
"eval_steps_per_second": 3.183,
|
778 |
+
"step": 55000
|
779 |
+
},
|
780 |
+
{
|
781 |
+
"epoch": 2.95,
|
782 |
+
"learning_rate": 3.637973304276764e-05,
|
783 |
+
"loss": 2.636,
|
784 |
+
"step": 56000
|
785 |
+
},
|
786 |
+
{
|
787 |
+
"epoch": 2.95,
|
788 |
+
"eval_loss": 5.104933261871338,
|
789 |
+
"eval_runtime": 81.9752,
|
790 |
+
"eval_samples_per_second": 50.918,
|
791 |
+
"eval_steps_per_second": 3.184,
|
792 |
+
"step": 56000
|
793 |
+
},
|
794 |
+
{
|
795 |
+
"epoch": 3.01,
|
796 |
+
"learning_rate": 3.6107327703623e-05,
|
797 |
+
"loss": 2.6014,
|
798 |
+
"step": 57000
|
799 |
+
},
|
800 |
+
{
|
801 |
+
"epoch": 3.01,
|
802 |
+
"eval_loss": 5.3848161697387695,
|
803 |
+
"eval_runtime": 80.9116,
|
804 |
+
"eval_samples_per_second": 51.587,
|
805 |
+
"eval_steps_per_second": 3.226,
|
806 |
+
"step": 57000
|
807 |
+
},
|
808 |
+
{
|
809 |
+
"epoch": 3.06,
|
810 |
+
"learning_rate": 3.583492236447834e-05,
|
811 |
+
"loss": 2.4145,
|
812 |
+
"step": 58000
|
813 |
+
},
|
814 |
+
{
|
815 |
+
"epoch": 3.06,
|
816 |
+
"eval_loss": 5.649404048919678,
|
817 |
+
"eval_runtime": 78.1589,
|
818 |
+
"eval_samples_per_second": 53.404,
|
819 |
+
"eval_steps_per_second": 3.339,
|
820 |
+
"step": 58000
|
821 |
+
},
|
822 |
+
{
|
823 |
+
"epoch": 3.11,
|
824 |
+
"learning_rate": 3.5562517025333695e-05,
|
825 |
+
"loss": 2.4236,
|
826 |
+
"step": 59000
|
827 |
+
},
|
828 |
+
{
|
829 |
+
"epoch": 3.11,
|
830 |
+
"eval_loss": 5.877135276794434,
|
831 |
+
"eval_runtime": 78.2014,
|
832 |
+
"eval_samples_per_second": 53.375,
|
833 |
+
"eval_steps_per_second": 3.338,
|
834 |
+
"step": 59000
|
835 |
+
},
|
836 |
+
{
|
837 |
+
"epoch": 3.17,
|
838 |
+
"learning_rate": 3.5290111686189054e-05,
|
839 |
+
"loss": 2.448,
|
840 |
+
"step": 60000
|
841 |
+
},
|
842 |
+
{
|
843 |
+
"epoch": 3.17,
|
844 |
+
"eval_loss": 5.5254974365234375,
|
845 |
+
"eval_runtime": 78.286,
|
846 |
+
"eval_samples_per_second": 53.317,
|
847 |
+
"eval_steps_per_second": 3.334,
|
848 |
+
"step": 60000
|
849 |
+
},
|
850 |
+
{
|
851 |
+
"epoch": 3.22,
|
852 |
+
"learning_rate": 3.5017706347044406e-05,
|
853 |
+
"loss": 2.4516,
|
854 |
+
"step": 61000
|
855 |
+
},
|
856 |
+
{
|
857 |
+
"epoch": 3.22,
|
858 |
+
"eval_loss": 5.600991249084473,
|
859 |
+
"eval_runtime": 78.1753,
|
860 |
+
"eval_samples_per_second": 53.393,
|
861 |
+
"eval_steps_per_second": 3.339,
|
862 |
+
"step": 61000
|
863 |
+
},
|
864 |
+
{
|
865 |
+
"epoch": 3.27,
|
866 |
+
"learning_rate": 3.474530100789976e-05,
|
867 |
+
"loss": 2.4536,
|
868 |
+
"step": 62000
|
869 |
+
},
|
870 |
+
{
|
871 |
+
"epoch": 3.27,
|
872 |
+
"eval_loss": 5.542355537414551,
|
873 |
+
"eval_runtime": 78.3728,
|
874 |
+
"eval_samples_per_second": 53.258,
|
875 |
+
"eval_steps_per_second": 3.33,
|
876 |
+
"step": 62000
|
877 |
+
},
|
878 |
+
{
|
879 |
+
"epoch": 3.32,
|
880 |
+
"learning_rate": 3.447289566875511e-05,
|
881 |
+
"loss": 2.4595,
|
882 |
+
"step": 63000
|
883 |
+
},
|
884 |
+
{
|
885 |
+
"epoch": 3.32,
|
886 |
+
"eval_loss": 5.826052188873291,
|
887 |
+
"eval_runtime": 78.1666,
|
888 |
+
"eval_samples_per_second": 53.399,
|
889 |
+
"eval_steps_per_second": 3.339,
|
890 |
+
"step": 63000
|
891 |
+
},
|
892 |
+
{
|
893 |
+
"epoch": 3.38,
|
894 |
+
"learning_rate": 3.420049032961046e-05,
|
895 |
+
"loss": 2.4539,
|
896 |
+
"step": 64000
|
897 |
+
},
|
898 |
+
{
|
899 |
+
"epoch": 3.38,
|
900 |
+
"eval_loss": 5.755754470825195,
|
901 |
+
"eval_runtime": 78.124,
|
902 |
+
"eval_samples_per_second": 53.428,
|
903 |
+
"eval_steps_per_second": 3.341,
|
904 |
+
"step": 64000
|
905 |
+
},
|
906 |
+
{
|
907 |
+
"epoch": 3.43,
|
908 |
+
"learning_rate": 3.3928084990465814e-05,
|
909 |
+
"loss": 2.4511,
|
910 |
+
"step": 65000
|
911 |
+
},
|
912 |
+
{
|
913 |
+
"epoch": 3.43,
|
914 |
+
"eval_loss": 5.6875505447387695,
|
915 |
+
"eval_runtime": 78.2257,
|
916 |
+
"eval_samples_per_second": 53.358,
|
917 |
+
"eval_steps_per_second": 3.336,
|
918 |
+
"step": 65000
|
919 |
+
},
|
920 |
+
{
|
921 |
+
"epoch": 3.48,
|
922 |
+
"learning_rate": 3.3655679651321166e-05,
|
923 |
+
"loss": 2.463,
|
924 |
+
"step": 66000
|
925 |
+
},
|
926 |
+
{
|
927 |
+
"epoch": 3.48,
|
928 |
+
"eval_loss": 5.699981689453125,
|
929 |
+
"eval_runtime": 78.274,
|
930 |
+
"eval_samples_per_second": 53.325,
|
931 |
+
"eval_steps_per_second": 3.334,
|
932 |
+
"step": 66000
|
933 |
+
},
|
934 |
+
{
|
935 |
+
"epoch": 3.53,
|
936 |
+
"learning_rate": 3.338327431217652e-05,
|
937 |
+
"loss": 2.4678,
|
938 |
+
"step": 67000
|
939 |
+
},
|
940 |
+
{
|
941 |
+
"epoch": 3.53,
|
942 |
+
"eval_loss": 5.762045383453369,
|
943 |
+
"eval_runtime": 78.2031,
|
944 |
+
"eval_samples_per_second": 53.374,
|
945 |
+
"eval_steps_per_second": 3.337,
|
946 |
+
"step": 67000
|
947 |
+
},
|
948 |
+
{
|
949 |
+
"epoch": 3.59,
|
950 |
+
"learning_rate": 3.311086897303187e-05,
|
951 |
+
"loss": 2.4753,
|
952 |
+
"step": 68000
|
953 |
+
},
|
954 |
+
{
|
955 |
+
"epoch": 3.59,
|
956 |
+
"eval_loss": 5.77708101272583,
|
957 |
+
"eval_runtime": 78.2487,
|
958 |
+
"eval_samples_per_second": 53.343,
|
959 |
+
"eval_steps_per_second": 3.336,
|
960 |
+
"step": 68000
|
961 |
+
},
|
962 |
+
{
|
963 |
+
"epoch": 3.64,
|
964 |
+
"learning_rate": 3.283846363388722e-05,
|
965 |
+
"loss": 2.4713,
|
966 |
+
"step": 69000
|
967 |
+
},
|
968 |
+
{
|
969 |
+
"epoch": 3.64,
|
970 |
+
"eval_loss": 5.61986780166626,
|
971 |
+
"eval_runtime": 78.1013,
|
972 |
+
"eval_samples_per_second": 53.443,
|
973 |
+
"eval_steps_per_second": 3.342,
|
974 |
+
"step": 69000
|
975 |
+
},
|
976 |
+
{
|
977 |
+
"epoch": 3.69,
|
978 |
+
"learning_rate": 3.256605829474258e-05,
|
979 |
+
"loss": 2.5024,
|
980 |
+
"step": 70000
|
981 |
+
},
|
982 |
+
{
|
983 |
+
"epoch": 3.69,
|
984 |
+
"eval_loss": 5.585144996643066,
|
985 |
+
"eval_runtime": 78.2816,
|
986 |
+
"eval_samples_per_second": 53.32,
|
987 |
+
"eval_steps_per_second": 3.334,
|
988 |
+
"step": 70000
|
989 |
+
},
|
990 |
+
{
|
991 |
+
"epoch": 3.75,
|
992 |
+
"learning_rate": 3.2293652955597933e-05,
|
993 |
+
"loss": 2.4772,
|
994 |
+
"step": 71000
|
995 |
+
},
|
996 |
+
{
|
997 |
+
"epoch": 3.75,
|
998 |
+
"eval_loss": 5.567023277282715,
|
999 |
+
"eval_runtime": 78.3144,
|
1000 |
+
"eval_samples_per_second": 53.298,
|
1001 |
+
"eval_steps_per_second": 3.333,
|
1002 |
+
"step": 71000
|
1003 |
+
},
|
1004 |
+
{
|
1005 |
+
"epoch": 3.8,
|
1006 |
+
"learning_rate": 3.202124761645328e-05,
|
1007 |
+
"loss": 2.4989,
|
1008 |
+
"step": 72000
|
1009 |
+
},
|
1010 |
+
{
|
1011 |
+
"epoch": 3.8,
|
1012 |
+
"eval_loss": 5.388617038726807,
|
1013 |
+
"eval_runtime": 78.311,
|
1014 |
+
"eval_samples_per_second": 53.3,
|
1015 |
+
"eval_steps_per_second": 3.333,
|
1016 |
+
"step": 72000
|
1017 |
+
},
|
1018 |
+
{
|
1019 |
+
"epoch": 3.85,
|
1020 |
+
"learning_rate": 3.174884227730864e-05,
|
1021 |
+
"loss": 2.4908,
|
1022 |
+
"step": 73000
|
1023 |
+
},
|
1024 |
+
{
|
1025 |
+
"epoch": 3.85,
|
1026 |
+
"eval_loss": 5.46661901473999,
|
1027 |
+
"eval_runtime": 78.3042,
|
1028 |
+
"eval_samples_per_second": 53.305,
|
1029 |
+
"eval_steps_per_second": 3.333,
|
1030 |
+
"step": 73000
|
1031 |
+
},
|
1032 |
+
{
|
1033 |
+
"epoch": 3.9,
|
1034 |
+
"learning_rate": 3.147643693816399e-05,
|
1035 |
+
"loss": 2.4975,
|
1036 |
+
"step": 74000
|
1037 |
+
},
|
1038 |
+
{
|
1039 |
+
"epoch": 3.9,
|
1040 |
+
"eval_loss": 5.72676944732666,
|
1041 |
+
"eval_runtime": 78.4025,
|
1042 |
+
"eval_samples_per_second": 53.238,
|
1043 |
+
"eval_steps_per_second": 3.329,
|
1044 |
+
"step": 74000
|
1045 |
+
},
|
1046 |
+
{
|
1047 |
+
"epoch": 3.96,
|
1048 |
+
"learning_rate": 3.120403159901934e-05,
|
1049 |
+
"loss": 2.4983,
|
1050 |
+
"step": 75000
|
1051 |
+
},
|
1052 |
+
{
|
1053 |
+
"epoch": 3.96,
|
1054 |
+
"eval_loss": 5.200565814971924,
|
1055 |
+
"eval_runtime": 78.2961,
|
1056 |
+
"eval_samples_per_second": 53.31,
|
1057 |
+
"eval_steps_per_second": 3.333,
|
1058 |
+
"step": 75000
|
1059 |
+
},
|
1060 |
+
{
|
1061 |
+
"epoch": 4.01,
|
1062 |
+
"learning_rate": 3.0931626259874694e-05,
|
1063 |
+
"loss": 2.4595,
|
1064 |
+
"step": 76000
|
1065 |
+
},
|
1066 |
+
{
|
1067 |
+
"epoch": 4.01,
|
1068 |
+
"eval_loss": 5.285388946533203,
|
1069 |
+
"eval_runtime": 77.5634,
|
1070 |
+
"eval_samples_per_second": 53.814,
|
1071 |
+
"eval_steps_per_second": 3.365,
|
1072 |
+
"step": 76000
|
1073 |
+
},
|
1074 |
+
{
|
1075 |
+
"epoch": 4.06,
|
1076 |
+
"learning_rate": 3.0659220920730046e-05,
|
1077 |
+
"loss": 2.2848,
|
1078 |
+
"step": 77000
|
1079 |
+
},
|
1080 |
+
{
|
1081 |
+
"epoch": 4.06,
|
1082 |
+
"eval_loss": 5.081736087799072,
|
1083 |
+
"eval_runtime": 77.374,
|
1084 |
+
"eval_samples_per_second": 53.946,
|
1085 |
+
"eval_steps_per_second": 3.373,
|
1086 |
+
"step": 77000
|
1087 |
+
},
|
1088 |
+
{
|
1089 |
+
"epoch": 4.12,
|
1090 |
+
"learning_rate": 3.0386815581585398e-05,
|
1091 |
+
"loss": 2.3182,
|
1092 |
+
"step": 78000
|
1093 |
+
},
|
1094 |
+
{
|
1095 |
+
"epoch": 4.12,
|
1096 |
+
"eval_loss": 5.363966464996338,
|
1097 |
+
"eval_runtime": 77.4691,
|
1098 |
+
"eval_samples_per_second": 53.88,
|
1099 |
+
"eval_steps_per_second": 3.369,
|
1100 |
+
"step": 78000
|
1101 |
+
},
|
1102 |
+
{
|
1103 |
+
"epoch": 4.17,
|
1104 |
+
"learning_rate": 3.0114410242440753e-05,
|
1105 |
+
"loss": 2.2998,
|
1106 |
+
"step": 79000
|
1107 |
+
},
|
1108 |
+
{
|
1109 |
+
"epoch": 4.17,
|
1110 |
+
"eval_loss": 5.359724044799805,
|
1111 |
+
"eval_runtime": 77.9931,
|
1112 |
+
"eval_samples_per_second": 53.518,
|
1113 |
+
"eval_steps_per_second": 3.346,
|
1114 |
+
"step": 79000
|
1115 |
+
},
|
1116 |
+
{
|
1117 |
+
"epoch": 4.22,
|
1118 |
+
"learning_rate": 2.9842004903296105e-05,
|
1119 |
+
"loss": 2.3085,
|
1120 |
+
"step": 80000
|
1121 |
+
},
|
1122 |
+
{
|
1123 |
+
"epoch": 4.22,
|
1124 |
+
"eval_loss": 5.6501922607421875,
|
1125 |
+
"eval_runtime": 77.8375,
|
1126 |
+
"eval_samples_per_second": 53.625,
|
1127 |
+
"eval_steps_per_second": 3.353,
|
1128 |
+
"step": 80000
|
1129 |
+
},
|
1130 |
+
{
|
1131 |
+
"epoch": 4.27,
|
1132 |
+
"learning_rate": 2.9569599564151457e-05,
|
1133 |
+
"loss": 2.2716,
|
1134 |
+
"step": 81000
|
1135 |
+
},
|
1136 |
+
{
|
1137 |
+
"epoch": 4.27,
|
1138 |
+
"eval_loss": 5.682339668273926,
|
1139 |
+
"eval_runtime": 78.1512,
|
1140 |
+
"eval_samples_per_second": 53.409,
|
1141 |
+
"eval_steps_per_second": 3.34,
|
1142 |
+
"step": 81000
|
1143 |
+
},
|
1144 |
+
{
|
1145 |
+
"epoch": 4.33,
|
1146 |
+
"learning_rate": 2.9297194225006813e-05,
|
1147 |
+
"loss": 2.2716,
|
1148 |
+
"step": 82000
|
1149 |
+
},
|
1150 |
+
{
|
1151 |
+
"epoch": 4.33,
|
1152 |
+
"eval_loss": 5.677391052246094,
|
1153 |
+
"eval_runtime": 78.1619,
|
1154 |
+
"eval_samples_per_second": 53.402,
|
1155 |
+
"eval_steps_per_second": 3.339,
|
1156 |
+
"step": 82000
|
1157 |
+
},
|
1158 |
+
{
|
1159 |
+
"epoch": 4.38,
|
1160 |
+
"learning_rate": 2.9024788885862165e-05,
|
1161 |
+
"loss": 2.284,
|
1162 |
+
"step": 83000
|
1163 |
+
},
|
1164 |
+
{
|
1165 |
+
"epoch": 4.38,
|
1166 |
+
"eval_loss": 5.525763988494873,
|
1167 |
+
"eval_runtime": 78.1139,
|
1168 |
+
"eval_samples_per_second": 53.435,
|
1169 |
+
"eval_steps_per_second": 3.341,
|
1170 |
+
"step": 83000
|
1171 |
+
},
|
1172 |
+
{
|
1173 |
+
"epoch": 4.43,
|
1174 |
+
"learning_rate": 2.8752383546717514e-05,
|
1175 |
+
"loss": 2.2916,
|
1176 |
+
"step": 84000
|
1177 |
+
},
|
1178 |
+
{
|
1179 |
+
"epoch": 4.43,
|
1180 |
+
"eval_loss": 5.277717590332031,
|
1181 |
+
"eval_runtime": 78.1112,
|
1182 |
+
"eval_samples_per_second": 53.437,
|
1183 |
+
"eval_steps_per_second": 3.341,
|
1184 |
+
"step": 84000
|
1185 |
+
},
|
1186 |
+
{
|
1187 |
+
"epoch": 4.48,
|
1188 |
+
"learning_rate": 2.8479978207572872e-05,
|
1189 |
+
"loss": 2.2978,
|
1190 |
+
"step": 85000
|
1191 |
+
},
|
1192 |
+
{
|
1193 |
+
"epoch": 4.48,
|
1194 |
+
"eval_loss": 5.4305100440979,
|
1195 |
+
"eval_runtime": 78.0088,
|
1196 |
+
"eval_samples_per_second": 53.507,
|
1197 |
+
"eval_steps_per_second": 3.346,
|
1198 |
+
"step": 85000
|
1199 |
+
},
|
1200 |
+
{
|
1201 |
+
"epoch": 4.54,
|
1202 |
+
"learning_rate": 2.820757286842822e-05,
|
1203 |
+
"loss": 2.3053,
|
1204 |
+
"step": 86000
|
1205 |
+
},
|
1206 |
+
{
|
1207 |
+
"epoch": 4.54,
|
1208 |
+
"eval_loss": 5.21947717666626,
|
1209 |
+
"eval_runtime": 77.9992,
|
1210 |
+
"eval_samples_per_second": 53.513,
|
1211 |
+
"eval_steps_per_second": 3.346,
|
1212 |
+
"step": 86000
|
1213 |
+
},
|
1214 |
+
{
|
1215 |
+
"epoch": 4.59,
|
1216 |
+
"learning_rate": 2.7935167529283573e-05,
|
1217 |
+
"loss": 2.3057,
|
1218 |
+
"step": 87000
|
1219 |
+
},
|
1220 |
+
{
|
1221 |
+
"epoch": 4.59,
|
1222 |
+
"eval_loss": 5.573265075683594,
|
1223 |
+
"eval_runtime": 78.0097,
|
1224 |
+
"eval_samples_per_second": 53.506,
|
1225 |
+
"eval_steps_per_second": 3.346,
|
1226 |
+
"step": 87000
|
1227 |
+
},
|
1228 |
+
{
|
1229 |
+
"epoch": 4.64,
|
1230 |
+
"learning_rate": 2.766276219013893e-05,
|
1231 |
+
"loss": 2.321,
|
1232 |
+
"step": 88000
|
1233 |
+
},
|
1234 |
+
{
|
1235 |
+
"epoch": 4.64,
|
1236 |
+
"eval_loss": 5.401345252990723,
|
1237 |
+
"eval_runtime": 77.9848,
|
1238 |
+
"eval_samples_per_second": 53.523,
|
1239 |
+
"eval_steps_per_second": 3.347,
|
1240 |
+
"step": 88000
|
1241 |
+
},
|
1242 |
+
{
|
1243 |
+
"epoch": 4.7,
|
1244 |
+
"learning_rate": 2.739035685099428e-05,
|
1245 |
+
"loss": 2.3067,
|
1246 |
+
"step": 89000
|
1247 |
+
},
|
1248 |
+
{
|
1249 |
+
"epoch": 4.7,
|
1250 |
+
"eval_loss": 5.604618072509766,
|
1251 |
+
"eval_runtime": 77.7786,
|
1252 |
+
"eval_samples_per_second": 53.665,
|
1253 |
+
"eval_steps_per_second": 3.356,
|
1254 |
+
"step": 89000
|
1255 |
+
},
|
1256 |
+
{
|
1257 |
+
"epoch": 4.75,
|
1258 |
+
"learning_rate": 2.7117951511849633e-05,
|
1259 |
+
"loss": 2.317,
|
1260 |
+
"step": 90000
|
1261 |
+
},
|
1262 |
+
{
|
1263 |
+
"epoch": 4.75,
|
1264 |
+
"eval_loss": 5.823883533477783,
|
1265 |
+
"eval_runtime": 77.9289,
|
1266 |
+
"eval_samples_per_second": 53.562,
|
1267 |
+
"eval_steps_per_second": 3.349,
|
1268 |
+
"step": 90000
|
1269 |
+
},
|
1270 |
+
{
|
1271 |
+
"epoch": 4.8,
|
1272 |
+
"learning_rate": 2.6845546172704988e-05,
|
1273 |
+
"loss": 2.3356,
|
1274 |
+
"step": 91000
|
1275 |
+
},
|
1276 |
+
{
|
1277 |
+
"epoch": 4.8,
|
1278 |
+
"eval_loss": 5.789203643798828,
|
1279 |
+
"eval_runtime": 78.1796,
|
1280 |
+
"eval_samples_per_second": 53.39,
|
1281 |
+
"eval_steps_per_second": 3.338,
|
1282 |
+
"step": 91000
|
1283 |
+
},
|
1284 |
+
{
|
1285 |
+
"epoch": 4.85,
|
1286 |
+
"learning_rate": 2.657314083356034e-05,
|
1287 |
+
"loss": 2.3301,
|
1288 |
+
"step": 92000
|
1289 |
+
},
|
1290 |
+
{
|
1291 |
+
"epoch": 4.85,
|
1292 |
+
"eval_loss": 5.643290996551514,
|
1293 |
+
"eval_runtime": 78.4214,
|
1294 |
+
"eval_samples_per_second": 53.225,
|
1295 |
+
"eval_steps_per_second": 3.328,
|
1296 |
+
"step": 92000
|
1297 |
+
},
|
1298 |
+
{
|
1299 |
+
"epoch": 4.91,
|
1300 |
+
"learning_rate": 2.630073549441569e-05,
|
1301 |
+
"loss": 2.3475,
|
1302 |
+
"step": 93000
|
1303 |
+
},
|
1304 |
+
{
|
1305 |
+
"epoch": 4.91,
|
1306 |
+
"eval_loss": 5.742900848388672,
|
1307 |
+
"eval_runtime": 78.38,
|
1308 |
+
"eval_samples_per_second": 53.253,
|
1309 |
+
"eval_steps_per_second": 3.33,
|
1310 |
+
"step": 93000
|
1311 |
+
},
|
1312 |
+
{
|
1313 |
+
"epoch": 4.96,
|
1314 |
+
"learning_rate": 2.6028330155271048e-05,
|
1315 |
+
"loss": 2.3237,
|
1316 |
+
"step": 94000
|
1317 |
+
},
|
1318 |
+
{
|
1319 |
+
"epoch": 4.96,
|
1320 |
+
"eval_loss": 5.759471416473389,
|
1321 |
+
"eval_runtime": 78.3893,
|
1322 |
+
"eval_samples_per_second": 53.247,
|
1323 |
+
"eval_steps_per_second": 3.33,
|
1324 |
+
"step": 94000
|
1325 |
+
},
|
1326 |
+
{
|
1327 |
+
"epoch": 5.01,
|
1328 |
+
"learning_rate": 2.5755924816126396e-05,
|
1329 |
+
"loss": 2.2959,
|
1330 |
+
"step": 95000
|
1331 |
+
},
|
1332 |
+
{
|
1333 |
+
"epoch": 5.01,
|
1334 |
+
"eval_loss": 5.965908050537109,
|
1335 |
+
"eval_runtime": 78.3426,
|
1336 |
+
"eval_samples_per_second": 53.279,
|
1337 |
+
"eval_steps_per_second": 3.332,
|
1338 |
+
"step": 95000
|
1339 |
+
},
|
1340 |
+
{
|
1341 |
+
"epoch": 5.06,
|
1342 |
+
"learning_rate": 2.548351947698175e-05,
|
1343 |
+
"loss": 2.192,
|
1344 |
+
"step": 96000
|
1345 |
+
},
|
1346 |
+
{
|
1347 |
+
"epoch": 5.06,
|
1348 |
+
"eval_loss": 5.630726337432861,
|
1349 |
+
"eval_runtime": 78.3297,
|
1350 |
+
"eval_samples_per_second": 53.288,
|
1351 |
+
"eval_steps_per_second": 3.332,
|
1352 |
+
"step": 96000
|
1353 |
+
},
|
1354 |
+
{
|
1355 |
+
"epoch": 5.12,
|
1356 |
+
"learning_rate": 2.5211114137837104e-05,
|
1357 |
+
"loss": 2.1869,
|
1358 |
+
"step": 97000
|
1359 |
+
},
|
1360 |
+
{
|
1361 |
+
"epoch": 5.12,
|
1362 |
+
"eval_loss": 5.92326545715332,
|
1363 |
+
"eval_runtime": 78.461,
|
1364 |
+
"eval_samples_per_second": 53.198,
|
1365 |
+
"eval_steps_per_second": 3.326,
|
1366 |
+
"step": 97000
|
1367 |
+
},
|
1368 |
+
{
|
1369 |
+
"epoch": 5.17,
|
1370 |
+
"learning_rate": 2.4938708798692456e-05,
|
1371 |
+
"loss": 2.203,
|
1372 |
+
"step": 98000
|
1373 |
+
},
|
1374 |
+
{
|
1375 |
+
"epoch": 5.17,
|
1376 |
+
"eval_loss": 5.989378929138184,
|
1377 |
+
"eval_runtime": 78.6971,
|
1378 |
+
"eval_samples_per_second": 53.039,
|
1379 |
+
"eval_steps_per_second": 3.317,
|
1380 |
+
"step": 98000
|
1381 |
+
},
|
1382 |
+
{
|
1383 |
+
"epoch": 5.22,
|
1384 |
+
"learning_rate": 2.4666303459547808e-05,
|
1385 |
+
"loss": 2.2222,
|
1386 |
+
"step": 99000
|
1387 |
+
},
|
1388 |
+
{
|
1389 |
+
"epoch": 5.22,
|
1390 |
+
"eval_loss": 5.885297775268555,
|
1391 |
+
"eval_runtime": 78.4475,
|
1392 |
+
"eval_samples_per_second": 53.208,
|
1393 |
+
"eval_steps_per_second": 3.327,
|
1394 |
+
"step": 99000
|
1395 |
+
},
|
1396 |
+
{
|
1397 |
+
"epoch": 5.28,
|
1398 |
+
"learning_rate": 2.439389812040316e-05,
|
1399 |
+
"loss": 2.225,
|
1400 |
+
"step": 100000
|
1401 |
+
},
|
1402 |
+
{
|
1403 |
+
"epoch": 5.28,
|
1404 |
+
"eval_loss": 5.9232964515686035,
|
1405 |
+
"eval_runtime": 78.1906,
|
1406 |
+
"eval_samples_per_second": 53.382,
|
1407 |
+
"eval_steps_per_second": 3.338,
|
1408 |
+
"step": 100000
|
1409 |
+
}
|
1410 |
+
],
|
1411 |
+
"max_steps": 189550,
|
1412 |
+
"num_train_epochs": 10,
|
1413 |
+
"total_flos": 2.5166251732608e+16,
|
1414 |
+
"trial_name": null,
|
1415 |
+
"trial_params": null
|
1416 |
+
}
|
training_args.bin
ADDED
Binary file (2.99 kB). View file
|
|