DorinSht commited on
Commit
e5c650f
1 Parent(s): cc45284

Training in progress, step 5000, checkpoint

Browse files
checkpoint-5000/config.json ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "JackFram/llama-68m",
3
+ "architectures": [
4
+ "LlamaForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 0,
9
+ "eos_token_id": 2,
10
+ "hidden_act": "silu",
11
+ "hidden_size": 768,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 3072,
14
+ "max_position_embeddings": 2048,
15
+ "model_type": "llama",
16
+ "num_attention_heads": 12,
17
+ "num_hidden_layers": 2,
18
+ "num_key_value_heads": 12,
19
+ "pad_token_id": 1,
20
+ "pretraining_tp": 1,
21
+ "rms_norm_eps": 1e-06,
22
+ "rope_scaling": null,
23
+ "rope_theta": 10000.0,
24
+ "tie_word_embeddings": false,
25
+ "torch_dtype": "float32",
26
+ "transformers_version": "4.41.0.dev0",
27
+ "use_cache": true,
28
+ "vocab_size": 32000
29
+ }
checkpoint-5000/generation_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 0,
4
+ "eos_token_id": 2,
5
+ "pad_token_id": 1,
6
+ "transformers_version": "4.41.0.dev0"
7
+ }
checkpoint-5000/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:585634ce4f342ad753db86ae66cc61f2ae52e3ecec0848257ffbc992fa5742f1
3
+ size 272123144
checkpoint-5000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:45914fc1ac6b561820396ff3ffd617f3383e294c0e0d217f6d148b077603d16d
3
+ size 544259743
checkpoint-5000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:69f50a692634404f2eebb2eab9f456865957578d752987bc52d843ac2a774366
3
+ size 14244
checkpoint-5000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b28aaef70505ca6b32511e2b3ef99dd89499d8108fb42685eeb57b7a062aa16f
3
+ size 1000
checkpoint-5000/special_tokens_map.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": true,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "</s>",
11
+ "lstrip": false,
12
+ "normalized": true,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": "<unk>",
17
+ "unk_token": {
18
+ "content": "<unk>",
19
+ "lstrip": false,
20
+ "normalized": true,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ }
24
+ }
checkpoint-5000/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-5000/tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
3
+ size 499723
checkpoint-5000/tokenizer_config.json ADDED
@@ -0,0 +1,45 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "add_prefix_space": true,
5
+ "added_tokens_decoder": {
6
+ "0": {
7
+ "content": "<unk>",
8
+ "lstrip": false,
9
+ "normalized": true,
10
+ "rstrip": false,
11
+ "single_word": false,
12
+ "special": true
13
+ },
14
+ "1": {
15
+ "content": "<s>",
16
+ "lstrip": false,
17
+ "normalized": true,
18
+ "rstrip": false,
19
+ "single_word": false,
20
+ "special": true
21
+ },
22
+ "2": {
23
+ "content": "</s>",
24
+ "lstrip": false,
25
+ "normalized": true,
26
+ "rstrip": false,
27
+ "single_word": false,
28
+ "special": true
29
+ }
30
+ },
31
+ "bos_token": "<s>",
32
+ "clean_up_tokenization_spaces": false,
33
+ "eos_token": "</s>",
34
+ "legacy": true,
35
+ "model_max_length": 2048,
36
+ "pad_token": "<unk>",
37
+ "padding": "max_length",
38
+ "return_tensors": "pt",
39
+ "sp_model_kwargs": {},
40
+ "spaces_between_special_tokens": false,
41
+ "tokenizer_class": "LlamaTokenizer",
42
+ "unk_token": "<unk>",
43
+ "use_default_system_prompt": false,
44
+ "use_fast": true
45
+ }
checkpoint-5000/trainer_state.json ADDED
@@ -0,0 +1,148 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.3220518244315178,
5
+ "eval_steps": 1000,
6
+ "global_step": 5000,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.13220518244315177,
13
+ "grad_norm": 0.8546391725540161,
14
+ "learning_rate": 8.816009873931059e-05,
15
+ "loss": 5.1118,
16
+ "step": 500
17
+ },
18
+ {
19
+ "epoch": 0.26441036488630354,
20
+ "grad_norm": 0.8593688607215881,
21
+ "learning_rate": 9.59831475011252e-05,
22
+ "loss": 3.406,
23
+ "step": 1000
24
+ },
25
+ {
26
+ "epoch": 0.26441036488630354,
27
+ "eval_accuracy": 0.5035306174465283,
28
+ "eval_loss": 3.23445987701416,
29
+ "eval_runtime": 74.0522,
30
+ "eval_samples_per_second": 24.847,
31
+ "eval_steps_per_second": 0.527,
32
+ "step": 1000
33
+ },
34
+ {
35
+ "epoch": 0.3966155473294553,
36
+ "grad_norm": 0.9617258906364441,
37
+ "learning_rate": 9.134314230431938e-05,
38
+ "loss": 3.0005,
39
+ "step": 1500
40
+ },
41
+ {
42
+ "epoch": 0.5288207297726071,
43
+ "grad_norm": 0.8953185677528381,
44
+ "learning_rate": 8.670313710751356e-05,
45
+ "loss": 2.8119,
46
+ "step": 2000
47
+ },
48
+ {
49
+ "epoch": 0.5288207297726071,
50
+ "eval_accuracy": 0.5365118094348038,
51
+ "eval_loss": 2.821384906768799,
52
+ "eval_runtime": 72.8065,
53
+ "eval_samples_per_second": 25.272,
54
+ "eval_steps_per_second": 0.536,
55
+ "step": 2000
56
+ },
57
+ {
58
+ "epoch": 0.6610259122157589,
59
+ "grad_norm": 1.4154396057128906,
60
+ "learning_rate": 8.206313191070773e-05,
61
+ "loss": 2.686,
62
+ "step": 2500
63
+ },
64
+ {
65
+ "epoch": 0.7932310946589106,
66
+ "grad_norm": 1.821349024772644,
67
+ "learning_rate": 7.742312671390191e-05,
68
+ "loss": 2.607,
69
+ "step": 3000
70
+ },
71
+ {
72
+ "epoch": 0.7932310946589106,
73
+ "eval_accuracy": 0.5497897240925214,
74
+ "eval_loss": 2.657219886779785,
75
+ "eval_runtime": 72.9553,
76
+ "eval_samples_per_second": 25.221,
77
+ "eval_steps_per_second": 0.535,
78
+ "step": 3000
79
+ },
80
+ {
81
+ "epoch": 0.9254362771020624,
82
+ "grad_norm": 2.0297396183013916,
83
+ "learning_rate": 7.278312151709609e-05,
84
+ "loss": 2.5642,
85
+ "step": 3500
86
+ },
87
+ {
88
+ "epoch": 1.0576414595452142,
89
+ "grad_norm": 2.8318285942077637,
90
+ "learning_rate": 6.814311632029027e-05,
91
+ "loss": 2.4734,
92
+ "step": 4000
93
+ },
94
+ {
95
+ "epoch": 1.0576414595452142,
96
+ "eval_accuracy": 0.5582058048894458,
97
+ "eval_loss": 2.5735702514648438,
98
+ "eval_runtime": 73.5253,
99
+ "eval_samples_per_second": 25.025,
100
+ "eval_steps_per_second": 0.53,
101
+ "step": 4000
102
+ },
103
+ {
104
+ "epoch": 1.189846641988366,
105
+ "grad_norm": 2.178349018096924,
106
+ "learning_rate": 6.350311112348446e-05,
107
+ "loss": 2.4542,
108
+ "step": 4500
109
+ },
110
+ {
111
+ "epoch": 1.3220518244315178,
112
+ "grad_norm": 2.944026470184326,
113
+ "learning_rate": 5.886310592667864e-05,
114
+ "loss": 2.4335,
115
+ "step": 5000
116
+ },
117
+ {
118
+ "epoch": 1.3220518244315178,
119
+ "eval_accuracy": 0.5617462458316518,
120
+ "eval_loss": 2.5335800647735596,
121
+ "eval_runtime": 73.7726,
122
+ "eval_samples_per_second": 24.941,
123
+ "eval_steps_per_second": 0.529,
124
+ "step": 5000
125
+ }
126
+ ],
127
+ "logging_steps": 500,
128
+ "max_steps": 11346,
129
+ "num_input_tokens_seen": 0,
130
+ "num_train_epochs": 3,
131
+ "save_steps": 1000,
132
+ "stateful_callbacks": {
133
+ "TrainerControl": {
134
+ "args": {
135
+ "should_epoch_stop": false,
136
+ "should_evaluate": false,
137
+ "should_log": false,
138
+ "should_save": true,
139
+ "should_training_stop": false
140
+ },
141
+ "attributes": {}
142
+ }
143
+ },
144
+ "total_flos": 6.406355574639821e+16,
145
+ "train_batch_size": 24,
146
+ "trial_name": null,
147
+ "trial_params": null
148
+ }
checkpoint-5000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:84a8c3b8deb94d253a20c7d637102edeaa129f75276f850abb181ddfd4b1ddcf
3
+ size 5176