Training in progress, epoch 2, checkpoint
Browse files- checkpoint-750/config.json +49 -0
- checkpoint-750/model.safetensors +3 -0
- checkpoint-750/optimizer.pt +3 -0
- checkpoint-750/rng_state.pth +3 -0
- checkpoint-750/scheduler.pt +3 -0
- checkpoint-750/special_tokens_map.json +51 -0
- checkpoint-750/tokenizer_config.json +75 -0
- checkpoint-750/trainer_state.json +487 -0
- checkpoint-750/training_args.bin +3 -0
- checkpoint-750/vocab.txt +0 -0
checkpoint-750/config.json
ADDED
@@ -0,0 +1,49 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "funnel-transformer/small-base",
|
3 |
+
"activation_dropout": 0.0,
|
4 |
+
"architectures": [
|
5 |
+
"FunnelForSequenceClassification"
|
6 |
+
],
|
7 |
+
"attention_dropout": 0.1,
|
8 |
+
"attention_type": "relative_shift",
|
9 |
+
"block_repeats": [
|
10 |
+
1,
|
11 |
+
1,
|
12 |
+
1
|
13 |
+
],
|
14 |
+
"block_sizes": [
|
15 |
+
4,
|
16 |
+
4,
|
17 |
+
4
|
18 |
+
],
|
19 |
+
"d_head": 64,
|
20 |
+
"d_inner": 3072,
|
21 |
+
"d_model": 768,
|
22 |
+
"hidden_act": "gelu_new",
|
23 |
+
"hidden_dropout": 0.1,
|
24 |
+
"id2label": {
|
25 |
+
"0": "HUMAN",
|
26 |
+
"1": "AI"
|
27 |
+
},
|
28 |
+
"initializer_range": 0.1,
|
29 |
+
"initializer_std": null,
|
30 |
+
"label2id": {
|
31 |
+
"AI": 1,
|
32 |
+
"HUMAN": 0
|
33 |
+
},
|
34 |
+
"layer_norm_eps": 1e-09,
|
35 |
+
"max_position_embeddings": 512,
|
36 |
+
"model_type": "funnel",
|
37 |
+
"n_head": 12,
|
38 |
+
"num_decoder_layers": 2,
|
39 |
+
"pool_q_only": true,
|
40 |
+
"pooling_type": "mean",
|
41 |
+
"problem_type": "single_label_classification",
|
42 |
+
"rel_attn_type": "factorized",
|
43 |
+
"separate_cls": true,
|
44 |
+
"torch_dtype": "float32",
|
45 |
+
"transformers_version": "4.35.2",
|
46 |
+
"truncate_seq": true,
|
47 |
+
"type_vocab_size": 3,
|
48 |
+
"vocab_size": 30522
|
49 |
+
}
|
checkpoint-750/model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ee82fa59d113fb4f520ba84e83ac30aac096a6ce6300ff8945dee78e3d50f037
|
3 |
+
size 464844376
|
checkpoint-750/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f2d7a09ffa2100d4e203499dcd97dfa1f23a3b0508a0412e96adb42fcc1659fa
|
3 |
+
size 929837114
|
checkpoint-750/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7e8b1733c2b8f101db2670af297df31070c1ec24eae20ad89a86584de4736313
|
3 |
+
size 14244
|
checkpoint-750/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5434c5f9e5ca8770e37021e68d9c7bb78ba37c8cef0b3377a36e71ccdbd5e9f0
|
3 |
+
size 1064
|
checkpoint-750/special_tokens_map.json
ADDED
@@ -0,0 +1,51 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"bos_token": {
|
3 |
+
"content": "<s>",
|
4 |
+
"lstrip": false,
|
5 |
+
"normalized": false,
|
6 |
+
"rstrip": false,
|
7 |
+
"single_word": false
|
8 |
+
},
|
9 |
+
"cls_token": {
|
10 |
+
"content": "<cls>",
|
11 |
+
"lstrip": false,
|
12 |
+
"normalized": false,
|
13 |
+
"rstrip": false,
|
14 |
+
"single_word": false
|
15 |
+
},
|
16 |
+
"eos_token": {
|
17 |
+
"content": "</s>",
|
18 |
+
"lstrip": false,
|
19 |
+
"normalized": false,
|
20 |
+
"rstrip": false,
|
21 |
+
"single_word": false
|
22 |
+
},
|
23 |
+
"mask_token": {
|
24 |
+
"content": "<mask>",
|
25 |
+
"lstrip": false,
|
26 |
+
"normalized": false,
|
27 |
+
"rstrip": false,
|
28 |
+
"single_word": false
|
29 |
+
},
|
30 |
+
"pad_token": {
|
31 |
+
"content": "<pad>",
|
32 |
+
"lstrip": false,
|
33 |
+
"normalized": false,
|
34 |
+
"rstrip": false,
|
35 |
+
"single_word": false
|
36 |
+
},
|
37 |
+
"sep_token": {
|
38 |
+
"content": "<sep>",
|
39 |
+
"lstrip": false,
|
40 |
+
"normalized": false,
|
41 |
+
"rstrip": false,
|
42 |
+
"single_word": false
|
43 |
+
},
|
44 |
+
"unk_token": {
|
45 |
+
"content": "<unk>",
|
46 |
+
"lstrip": false,
|
47 |
+
"normalized": false,
|
48 |
+
"rstrip": false,
|
49 |
+
"single_word": false
|
50 |
+
}
|
51 |
+
}
|
checkpoint-750/tokenizer_config.json
ADDED
@@ -0,0 +1,75 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"added_tokens_decoder": {
|
3 |
+
"0": {
|
4 |
+
"content": "<pad>",
|
5 |
+
"lstrip": false,
|
6 |
+
"normalized": false,
|
7 |
+
"rstrip": false,
|
8 |
+
"single_word": false,
|
9 |
+
"special": true
|
10 |
+
},
|
11 |
+
"96": {
|
12 |
+
"content": "<s>",
|
13 |
+
"lstrip": false,
|
14 |
+
"normalized": false,
|
15 |
+
"rstrip": false,
|
16 |
+
"single_word": false,
|
17 |
+
"special": true
|
18 |
+
},
|
19 |
+
"97": {
|
20 |
+
"content": "</s>",
|
21 |
+
"lstrip": false,
|
22 |
+
"normalized": false,
|
23 |
+
"rstrip": false,
|
24 |
+
"single_word": false,
|
25 |
+
"special": true
|
26 |
+
},
|
27 |
+
"100": {
|
28 |
+
"content": "<unk>",
|
29 |
+
"lstrip": false,
|
30 |
+
"normalized": false,
|
31 |
+
"rstrip": false,
|
32 |
+
"single_word": false,
|
33 |
+
"special": true
|
34 |
+
},
|
35 |
+
"101": {
|
36 |
+
"content": "<cls>",
|
37 |
+
"lstrip": false,
|
38 |
+
"normalized": false,
|
39 |
+
"rstrip": false,
|
40 |
+
"single_word": false,
|
41 |
+
"special": true
|
42 |
+
},
|
43 |
+
"102": {
|
44 |
+
"content": "<sep>",
|
45 |
+
"lstrip": false,
|
46 |
+
"normalized": false,
|
47 |
+
"rstrip": false,
|
48 |
+
"single_word": false,
|
49 |
+
"special": true
|
50 |
+
},
|
51 |
+
"103": {
|
52 |
+
"content": "<mask>",
|
53 |
+
"lstrip": false,
|
54 |
+
"normalized": false,
|
55 |
+
"rstrip": false,
|
56 |
+
"single_word": false,
|
57 |
+
"special": true
|
58 |
+
}
|
59 |
+
},
|
60 |
+
"bos_token": "<s>",
|
61 |
+
"clean_up_tokenization_spaces": true,
|
62 |
+
"cls_token": "<cls>",
|
63 |
+
"do_basic_tokenize": true,
|
64 |
+
"do_lower_case": true,
|
65 |
+
"eos_token": "</s>",
|
66 |
+
"mask_token": "<mask>",
|
67 |
+
"model_max_length": 512,
|
68 |
+
"never_split": null,
|
69 |
+
"pad_token": "<pad>",
|
70 |
+
"sep_token": "<sep>",
|
71 |
+
"strip_accents": null,
|
72 |
+
"tokenize_chinese_chars": true,
|
73 |
+
"tokenizer_class": "FunnelTokenizer",
|
74 |
+
"unk_token": "<unk>"
|
75 |
+
}
|
checkpoint-750/trainer_state.json
ADDED
@@ -0,0 +1,487 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 2.0,
|
5 |
+
"eval_steps": 375,
|
6 |
+
"global_step": 750,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.03,
|
13 |
+
"learning_rate": 9.866666666666668e-06,
|
14 |
+
"loss": 0.6931,
|
15 |
+
"step": 10
|
16 |
+
},
|
17 |
+
{
|
18 |
+
"epoch": 0.05,
|
19 |
+
"learning_rate": 9.733333333333334e-06,
|
20 |
+
"loss": 0.6765,
|
21 |
+
"step": 20
|
22 |
+
},
|
23 |
+
{
|
24 |
+
"epoch": 0.08,
|
25 |
+
"learning_rate": 9.600000000000001e-06,
|
26 |
+
"loss": 0.4544,
|
27 |
+
"step": 30
|
28 |
+
},
|
29 |
+
{
|
30 |
+
"epoch": 0.11,
|
31 |
+
"learning_rate": 9.48e-06,
|
32 |
+
"loss": 0.2259,
|
33 |
+
"step": 40
|
34 |
+
},
|
35 |
+
{
|
36 |
+
"epoch": 0.13,
|
37 |
+
"learning_rate": 9.346666666666666e-06,
|
38 |
+
"loss": 0.1407,
|
39 |
+
"step": 50
|
40 |
+
},
|
41 |
+
{
|
42 |
+
"epoch": 0.16,
|
43 |
+
"learning_rate": 9.213333333333334e-06,
|
44 |
+
"loss": 0.1243,
|
45 |
+
"step": 60
|
46 |
+
},
|
47 |
+
{
|
48 |
+
"epoch": 0.19,
|
49 |
+
"learning_rate": 9.080000000000001e-06,
|
50 |
+
"loss": 0.0641,
|
51 |
+
"step": 70
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.21,
|
55 |
+
"learning_rate": 8.946666666666669e-06,
|
56 |
+
"loss": 0.0083,
|
57 |
+
"step": 80
|
58 |
+
},
|
59 |
+
{
|
60 |
+
"epoch": 0.24,
|
61 |
+
"learning_rate": 8.813333333333334e-06,
|
62 |
+
"loss": 0.0066,
|
63 |
+
"step": 90
|
64 |
+
},
|
65 |
+
{
|
66 |
+
"epoch": 0.27,
|
67 |
+
"learning_rate": 8.693333333333334e-06,
|
68 |
+
"loss": 0.0596,
|
69 |
+
"step": 100
|
70 |
+
},
|
71 |
+
{
|
72 |
+
"epoch": 0.29,
|
73 |
+
"learning_rate": 8.560000000000001e-06,
|
74 |
+
"loss": 0.0792,
|
75 |
+
"step": 110
|
76 |
+
},
|
77 |
+
{
|
78 |
+
"epoch": 0.32,
|
79 |
+
"learning_rate": 8.44e-06,
|
80 |
+
"loss": 0.1432,
|
81 |
+
"step": 120
|
82 |
+
},
|
83 |
+
{
|
84 |
+
"epoch": 0.35,
|
85 |
+
"learning_rate": 8.306666666666668e-06,
|
86 |
+
"loss": 0.117,
|
87 |
+
"step": 130
|
88 |
+
},
|
89 |
+
{
|
90 |
+
"epoch": 0.37,
|
91 |
+
"learning_rate": 8.173333333333334e-06,
|
92 |
+
"loss": 0.2646,
|
93 |
+
"step": 140
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.4,
|
97 |
+
"learning_rate": 8.040000000000001e-06,
|
98 |
+
"loss": 0.0097,
|
99 |
+
"step": 150
|
100 |
+
},
|
101 |
+
{
|
102 |
+
"epoch": 0.43,
|
103 |
+
"learning_rate": 7.906666666666667e-06,
|
104 |
+
"loss": 0.0333,
|
105 |
+
"step": 160
|
106 |
+
},
|
107 |
+
{
|
108 |
+
"epoch": 0.45,
|
109 |
+
"learning_rate": 7.773333333333334e-06,
|
110 |
+
"loss": 0.0016,
|
111 |
+
"step": 170
|
112 |
+
},
|
113 |
+
{
|
114 |
+
"epoch": 0.48,
|
115 |
+
"learning_rate": 7.640000000000001e-06,
|
116 |
+
"loss": 0.0038,
|
117 |
+
"step": 180
|
118 |
+
},
|
119 |
+
{
|
120 |
+
"epoch": 0.51,
|
121 |
+
"learning_rate": 7.506666666666668e-06,
|
122 |
+
"loss": 0.074,
|
123 |
+
"step": 190
|
124 |
+
},
|
125 |
+
{
|
126 |
+
"epoch": 0.53,
|
127 |
+
"learning_rate": 7.373333333333334e-06,
|
128 |
+
"loss": 0.1338,
|
129 |
+
"step": 200
|
130 |
+
},
|
131 |
+
{
|
132 |
+
"epoch": 0.56,
|
133 |
+
"learning_rate": 7.24e-06,
|
134 |
+
"loss": 0.0035,
|
135 |
+
"step": 210
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.59,
|
139 |
+
"learning_rate": 7.1066666666666675e-06,
|
140 |
+
"loss": 0.1519,
|
141 |
+
"step": 220
|
142 |
+
},
|
143 |
+
{
|
144 |
+
"epoch": 0.61,
|
145 |
+
"learning_rate": 6.973333333333334e-06,
|
146 |
+
"loss": 0.0095,
|
147 |
+
"step": 230
|
148 |
+
},
|
149 |
+
{
|
150 |
+
"epoch": 0.64,
|
151 |
+
"learning_rate": 6.8400000000000014e-06,
|
152 |
+
"loss": 0.1465,
|
153 |
+
"step": 240
|
154 |
+
},
|
155 |
+
{
|
156 |
+
"epoch": 0.67,
|
157 |
+
"learning_rate": 6.706666666666667e-06,
|
158 |
+
"loss": 0.0022,
|
159 |
+
"step": 250
|
160 |
+
},
|
161 |
+
{
|
162 |
+
"epoch": 0.69,
|
163 |
+
"learning_rate": 6.573333333333334e-06,
|
164 |
+
"loss": 0.0551,
|
165 |
+
"step": 260
|
166 |
+
},
|
167 |
+
{
|
168 |
+
"epoch": 0.72,
|
169 |
+
"learning_rate": 6.440000000000001e-06,
|
170 |
+
"loss": 0.0028,
|
171 |
+
"step": 270
|
172 |
+
},
|
173 |
+
{
|
174 |
+
"epoch": 0.75,
|
175 |
+
"learning_rate": 6.3066666666666676e-06,
|
176 |
+
"loss": 0.0566,
|
177 |
+
"step": 280
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 0.77,
|
181 |
+
"learning_rate": 6.173333333333333e-06,
|
182 |
+
"loss": 0.0009,
|
183 |
+
"step": 290
|
184 |
+
},
|
185 |
+
{
|
186 |
+
"epoch": 0.8,
|
187 |
+
"learning_rate": 6.040000000000001e-06,
|
188 |
+
"loss": 0.0902,
|
189 |
+
"step": 300
|
190 |
+
},
|
191 |
+
{
|
192 |
+
"epoch": 0.83,
|
193 |
+
"learning_rate": 5.906666666666667e-06,
|
194 |
+
"loss": 0.0008,
|
195 |
+
"step": 310
|
196 |
+
},
|
197 |
+
{
|
198 |
+
"epoch": 0.85,
|
199 |
+
"learning_rate": 5.7733333333333345e-06,
|
200 |
+
"loss": 0.0011,
|
201 |
+
"step": 320
|
202 |
+
},
|
203 |
+
{
|
204 |
+
"epoch": 0.88,
|
205 |
+
"learning_rate": 5.64e-06,
|
206 |
+
"loss": 0.0196,
|
207 |
+
"step": 330
|
208 |
+
},
|
209 |
+
{
|
210 |
+
"epoch": 0.91,
|
211 |
+
"learning_rate": 5.506666666666667e-06,
|
212 |
+
"loss": 0.0826,
|
213 |
+
"step": 340
|
214 |
+
},
|
215 |
+
{
|
216 |
+
"epoch": 0.93,
|
217 |
+
"learning_rate": 5.373333333333334e-06,
|
218 |
+
"loss": 0.0408,
|
219 |
+
"step": 350
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 0.96,
|
223 |
+
"learning_rate": 5.240000000000001e-06,
|
224 |
+
"loss": 0.0466,
|
225 |
+
"step": 360
|
226 |
+
},
|
227 |
+
{
|
228 |
+
"epoch": 0.99,
|
229 |
+
"learning_rate": 5.106666666666667e-06,
|
230 |
+
"loss": 0.0013,
|
231 |
+
"step": 370
|
232 |
+
},
|
233 |
+
{
|
234 |
+
"epoch": 1.0,
|
235 |
+
"eval_accuracy": 0.992,
|
236 |
+
"eval_loss": 0.05685167387127876,
|
237 |
+
"eval_runtime": 14.5309,
|
238 |
+
"eval_samples_per_second": 68.819,
|
239 |
+
"eval_steps_per_second": 8.602,
|
240 |
+
"step": 375
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 1.01,
|
244 |
+
"learning_rate": 4.973333333333334e-06,
|
245 |
+
"loss": 0.0205,
|
246 |
+
"step": 380
|
247 |
+
},
|
248 |
+
{
|
249 |
+
"epoch": 1.04,
|
250 |
+
"learning_rate": 4.84e-06,
|
251 |
+
"loss": 0.0767,
|
252 |
+
"step": 390
|
253 |
+
},
|
254 |
+
{
|
255 |
+
"epoch": 1.07,
|
256 |
+
"learning_rate": 4.706666666666667e-06,
|
257 |
+
"loss": 0.1028,
|
258 |
+
"step": 400
|
259 |
+
},
|
260 |
+
{
|
261 |
+
"epoch": 1.09,
|
262 |
+
"learning_rate": 4.573333333333333e-06,
|
263 |
+
"loss": 0.0007,
|
264 |
+
"step": 410
|
265 |
+
},
|
266 |
+
{
|
267 |
+
"epoch": 1.12,
|
268 |
+
"learning_rate": 4.440000000000001e-06,
|
269 |
+
"loss": 0.0717,
|
270 |
+
"step": 420
|
271 |
+
},
|
272 |
+
{
|
273 |
+
"epoch": 1.15,
|
274 |
+
"learning_rate": 4.306666666666666e-06,
|
275 |
+
"loss": 0.0008,
|
276 |
+
"step": 430
|
277 |
+
},
|
278 |
+
{
|
279 |
+
"epoch": 1.17,
|
280 |
+
"learning_rate": 4.173333333333334e-06,
|
281 |
+
"loss": 0.0005,
|
282 |
+
"step": 440
|
283 |
+
},
|
284 |
+
{
|
285 |
+
"epoch": 1.2,
|
286 |
+
"learning_rate": 4.04e-06,
|
287 |
+
"loss": 0.0007,
|
288 |
+
"step": 450
|
289 |
+
},
|
290 |
+
{
|
291 |
+
"epoch": 1.23,
|
292 |
+
"learning_rate": 3.906666666666667e-06,
|
293 |
+
"loss": 0.0886,
|
294 |
+
"step": 460
|
295 |
+
},
|
296 |
+
{
|
297 |
+
"epoch": 1.25,
|
298 |
+
"learning_rate": 3.7733333333333338e-06,
|
299 |
+
"loss": 0.0413,
|
300 |
+
"step": 470
|
301 |
+
},
|
302 |
+
{
|
303 |
+
"epoch": 1.28,
|
304 |
+
"learning_rate": 3.6400000000000003e-06,
|
305 |
+
"loss": 0.0223,
|
306 |
+
"step": 480
|
307 |
+
},
|
308 |
+
{
|
309 |
+
"epoch": 1.31,
|
310 |
+
"learning_rate": 3.5066666666666673e-06,
|
311 |
+
"loss": 0.0004,
|
312 |
+
"step": 490
|
313 |
+
},
|
314 |
+
{
|
315 |
+
"epoch": 1.33,
|
316 |
+
"learning_rate": 3.3733333333333334e-06,
|
317 |
+
"loss": 0.0146,
|
318 |
+
"step": 500
|
319 |
+
},
|
320 |
+
{
|
321 |
+
"epoch": 1.36,
|
322 |
+
"learning_rate": 3.2400000000000003e-06,
|
323 |
+
"loss": 0.0007,
|
324 |
+
"step": 510
|
325 |
+
},
|
326 |
+
{
|
327 |
+
"epoch": 1.39,
|
328 |
+
"learning_rate": 3.106666666666667e-06,
|
329 |
+
"loss": 0.0201,
|
330 |
+
"step": 520
|
331 |
+
},
|
332 |
+
{
|
333 |
+
"epoch": 1.41,
|
334 |
+
"learning_rate": 2.973333333333334e-06,
|
335 |
+
"loss": 0.0019,
|
336 |
+
"step": 530
|
337 |
+
},
|
338 |
+
{
|
339 |
+
"epoch": 1.44,
|
340 |
+
"learning_rate": 2.84e-06,
|
341 |
+
"loss": 0.0035,
|
342 |
+
"step": 540
|
343 |
+
},
|
344 |
+
{
|
345 |
+
"epoch": 1.47,
|
346 |
+
"learning_rate": 2.706666666666667e-06,
|
347 |
+
"loss": 0.0126,
|
348 |
+
"step": 550
|
349 |
+
},
|
350 |
+
{
|
351 |
+
"epoch": 1.49,
|
352 |
+
"learning_rate": 2.573333333333334e-06,
|
353 |
+
"loss": 0.0004,
|
354 |
+
"step": 560
|
355 |
+
},
|
356 |
+
{
|
357 |
+
"epoch": 1.52,
|
358 |
+
"learning_rate": 2.4400000000000004e-06,
|
359 |
+
"loss": 0.0559,
|
360 |
+
"step": 570
|
361 |
+
},
|
362 |
+
{
|
363 |
+
"epoch": 1.55,
|
364 |
+
"learning_rate": 2.306666666666667e-06,
|
365 |
+
"loss": 0.0375,
|
366 |
+
"step": 580
|
367 |
+
},
|
368 |
+
{
|
369 |
+
"epoch": 1.57,
|
370 |
+
"learning_rate": 2.1733333333333334e-06,
|
371 |
+
"loss": 0.0004,
|
372 |
+
"step": 590
|
373 |
+
},
|
374 |
+
{
|
375 |
+
"epoch": 1.6,
|
376 |
+
"learning_rate": 2.04e-06,
|
377 |
+
"loss": 0.0003,
|
378 |
+
"step": 600
|
379 |
+
},
|
380 |
+
{
|
381 |
+
"epoch": 1.63,
|
382 |
+
"learning_rate": 1.906666666666667e-06,
|
383 |
+
"loss": 0.0004,
|
384 |
+
"step": 610
|
385 |
+
},
|
386 |
+
{
|
387 |
+
"epoch": 1.65,
|
388 |
+
"learning_rate": 1.7733333333333336e-06,
|
389 |
+
"loss": 0.0004,
|
390 |
+
"step": 620
|
391 |
+
},
|
392 |
+
{
|
393 |
+
"epoch": 1.68,
|
394 |
+
"learning_rate": 1.6400000000000002e-06,
|
395 |
+
"loss": 0.006,
|
396 |
+
"step": 630
|
397 |
+
},
|
398 |
+
{
|
399 |
+
"epoch": 1.71,
|
400 |
+
"learning_rate": 1.506666666666667e-06,
|
401 |
+
"loss": 0.0771,
|
402 |
+
"step": 640
|
403 |
+
},
|
404 |
+
{
|
405 |
+
"epoch": 1.73,
|
406 |
+
"learning_rate": 1.3733333333333335e-06,
|
407 |
+
"loss": 0.0003,
|
408 |
+
"step": 650
|
409 |
+
},
|
410 |
+
{
|
411 |
+
"epoch": 1.76,
|
412 |
+
"learning_rate": 1.2400000000000002e-06,
|
413 |
+
"loss": 0.096,
|
414 |
+
"step": 660
|
415 |
+
},
|
416 |
+
{
|
417 |
+
"epoch": 1.79,
|
418 |
+
"learning_rate": 1.1066666666666667e-06,
|
419 |
+
"loss": 0.0863,
|
420 |
+
"step": 670
|
421 |
+
},
|
422 |
+
{
|
423 |
+
"epoch": 1.81,
|
424 |
+
"learning_rate": 9.733333333333333e-07,
|
425 |
+
"loss": 0.0604,
|
426 |
+
"step": 680
|
427 |
+
},
|
428 |
+
{
|
429 |
+
"epoch": 1.84,
|
430 |
+
"learning_rate": 8.400000000000001e-07,
|
431 |
+
"loss": 0.0005,
|
432 |
+
"step": 690
|
433 |
+
},
|
434 |
+
{
|
435 |
+
"epoch": 1.87,
|
436 |
+
"learning_rate": 7.066666666666667e-07,
|
437 |
+
"loss": 0.0004,
|
438 |
+
"step": 700
|
439 |
+
},
|
440 |
+
{
|
441 |
+
"epoch": 1.89,
|
442 |
+
"learning_rate": 5.733333333333334e-07,
|
443 |
+
"loss": 0.1011,
|
444 |
+
"step": 710
|
445 |
+
},
|
446 |
+
{
|
447 |
+
"epoch": 1.92,
|
448 |
+
"learning_rate": 4.4e-07,
|
449 |
+
"loss": 0.0004,
|
450 |
+
"step": 720
|
451 |
+
},
|
452 |
+
{
|
453 |
+
"epoch": 1.95,
|
454 |
+
"learning_rate": 3.0666666666666666e-07,
|
455 |
+
"loss": 0.0003,
|
456 |
+
"step": 730
|
457 |
+
},
|
458 |
+
{
|
459 |
+
"epoch": 1.97,
|
460 |
+
"learning_rate": 1.7333333333333335e-07,
|
461 |
+
"loss": 0.0007,
|
462 |
+
"step": 740
|
463 |
+
},
|
464 |
+
{
|
465 |
+
"epoch": 2.0,
|
466 |
+
"learning_rate": 4e-08,
|
467 |
+
"loss": 0.0005,
|
468 |
+
"step": 750
|
469 |
+
},
|
470 |
+
{
|
471 |
+
"epoch": 2.0,
|
472 |
+
"eval_accuracy": 0.966,
|
473 |
+
"eval_loss": 0.17487600445747375,
|
474 |
+
"eval_runtime": 14.6356,
|
475 |
+
"eval_samples_per_second": 68.327,
|
476 |
+
"eval_steps_per_second": 8.541,
|
477 |
+
"step": 750
|
478 |
+
}
|
479 |
+
],
|
480 |
+
"logging_steps": 10,
|
481 |
+
"max_steps": 750,
|
482 |
+
"num_train_epochs": 2,
|
483 |
+
"save_steps": 500,
|
484 |
+
"total_flos": 1694524054894848.0,
|
485 |
+
"trial_name": null,
|
486 |
+
"trial_params": null
|
487 |
+
}
|
checkpoint-750/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c447d199ad29bb02a1c18507c2ef1baf344faf85ac1c60fba1abfb7baee9a5f1
|
3 |
+
size 4600
|
checkpoint-750/vocab.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|