yongzx commited on
Commit
d3ea3c8
1 Parent(s): cfd048e

Push to HF hub

Browse files
.gitattributes CHANGED
@@ -31,3 +31,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
31
  *.zip filter=lfs diff=lfs merge=lfs -text
32
  *.zst filter=lfs diff=lfs merge=lfs -text
33
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
31
  *.zip filter=lfs diff=lfs merge=lfs -text
32
  *.zst filter=lfs diff=lfs merge=lfs -text
33
  *tfevents* filter=lfs diff=lfs merge=lfs -text
34
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
.gitignore ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ checkpoint-20000/
2
+ checkpoint-25000/
3
+ pytorch_model-00001-of-00002.bin
4
+ pytorch_model-00002-of-00002.bin
5
+ checkpoint-15000/
6
+ pytorch_model.bin.index.json
7
+ checkpoint-10000/
8
+ checkpoint-5000/
config.json ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "bigscience/bloom-7b1",
3
+ "adapters": {
4
+ "adapters": {},
5
+ "config_map": {},
6
+ "fusion_config_map": {},
7
+ "fusions": {}
8
+ },
9
+ "apply_residual_connection_post_layernorm": false,
10
+ "architectures": [
11
+ "BloomForCausalLM"
12
+ ],
13
+ "attention_dropout": 0.0,
14
+ "attention_softmax_in_fp32": true,
15
+ "bias_dropout_fusion": true,
16
+ "bos_token_id": 1,
17
+ "eos_token_id": 2,
18
+ "hidden_dropout": 0.0,
19
+ "hidden_size": 4096,
20
+ "initializer_range": 0.02,
21
+ "layer_norm_epsilon": 1e-05,
22
+ "masked_softmax_fusion": true,
23
+ "model_type": "bloom",
24
+ "n_head": 32,
25
+ "n_inner": null,
26
+ "n_layer": 30,
27
+ "offset_alibi": 100,
28
+ "pad_token_id": 3,
29
+ "pretraining_tp": 1,
30
+ "skip_bias_add": true,
31
+ "skip_bias_add_qkv": false,
32
+ "slow_but_exact": false,
33
+ "torch_dtype": "float16",
34
+ "transformers_version": "4.20.0.dev0",
35
+ "unk_token_id": 0,
36
+ "use_cache": true,
37
+ "vocab_size": 250880
38
+ }
special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>"}
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8f6efc66e73f1fd69da4f436e48befb519fdff3fe18910850c1d41bd862293a5
3
+ size 14500443
tokenizer_config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"unk_token": "<unk>", "bos_token": "<s>", "eos_token": "</s>", "pad_token": "<pad>", "add_prefix_space": false, "name_or_path": "bigscience/bloom-7b1", "special_tokens_map_file": null, "padding_side": "left", "tokenizer_class": "BloomTokenizer"}
trainer_state.json ADDED
@@ -0,0 +1,116 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 1.6806640625,
3
+ "best_model_checkpoint": "/home/zhengxinyong/outputs/bloom-7b1_ru_continual-pretrain_100000samples_-1vocab_original/checkpoint-25000",
4
+ "epoch": 1.4637002341920375,
5
+ "global_step": 25000,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.15,
12
+ "learning_rate": 0.0001,
13
+ "loss": 2.1762,
14
+ "step": 2500
15
+ },
16
+ {
17
+ "epoch": 0.29,
18
+ "learning_rate": 0.0001,
19
+ "loss": 1.9292,
20
+ "step": 5000
21
+ },
22
+ {
23
+ "epoch": 0.29,
24
+ "eval_loss": 1.8681640625,
25
+ "eval_runtime": 20.6594,
26
+ "eval_samples_per_second": 65.975,
27
+ "eval_steps_per_second": 8.277,
28
+ "step": 5000
29
+ },
30
+ {
31
+ "epoch": 0.44,
32
+ "learning_rate": 0.0001,
33
+ "loss": 1.8402,
34
+ "step": 7500
35
+ },
36
+ {
37
+ "epoch": 0.59,
38
+ "learning_rate": 0.0001,
39
+ "loss": 1.8012,
40
+ "step": 10000
41
+ },
42
+ {
43
+ "epoch": 0.59,
44
+ "eval_loss": 1.7763671875,
45
+ "eval_runtime": 20.6347,
46
+ "eval_samples_per_second": 66.054,
47
+ "eval_steps_per_second": 8.287,
48
+ "step": 10000
49
+ },
50
+ {
51
+ "epoch": 0.73,
52
+ "learning_rate": 0.0001,
53
+ "loss": 1.7654,
54
+ "step": 12500
55
+ },
56
+ {
57
+ "epoch": 0.88,
58
+ "learning_rate": 0.0001,
59
+ "loss": 1.7347,
60
+ "step": 15000
61
+ },
62
+ {
63
+ "epoch": 0.88,
64
+ "eval_loss": 1.734375,
65
+ "eval_runtime": 20.6431,
66
+ "eval_samples_per_second": 66.027,
67
+ "eval_steps_per_second": 8.284,
68
+ "step": 15000
69
+ },
70
+ {
71
+ "epoch": 1.02,
72
+ "learning_rate": 0.0001,
73
+ "loss": 1.6922,
74
+ "step": 17500
75
+ },
76
+ {
77
+ "epoch": 1.17,
78
+ "learning_rate": 0.0001,
79
+ "loss": 1.5617,
80
+ "step": 20000
81
+ },
82
+ {
83
+ "epoch": 1.17,
84
+ "eval_loss": 1.703125,
85
+ "eval_runtime": 20.6213,
86
+ "eval_samples_per_second": 66.097,
87
+ "eval_steps_per_second": 8.292,
88
+ "step": 20000
89
+ },
90
+ {
91
+ "epoch": 1.32,
92
+ "learning_rate": 0.0001,
93
+ "loss": 1.5768,
94
+ "step": 22500
95
+ },
96
+ {
97
+ "epoch": 1.46,
98
+ "learning_rate": 0.0001,
99
+ "loss": 1.573,
100
+ "step": 25000
101
+ },
102
+ {
103
+ "epoch": 1.46,
104
+ "eval_loss": 1.6806640625,
105
+ "eval_runtime": 20.6399,
106
+ "eval_samples_per_second": 66.037,
107
+ "eval_steps_per_second": 8.285,
108
+ "step": 25000
109
+ }
110
+ ],
111
+ "max_steps": 25000,
112
+ "num_train_epochs": 2,
113
+ "total_flos": 7.423686731586601e+18,
114
+ "trial_name": null,
115
+ "trial_params": null
116
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a7fe5e7e0c83d8de028d678ceae44d37c6db24cb4d062c90fd6bbba822675b1e
3
+ size 4399