theblackcat102
commited on
Commit
•
6a85b57
1
Parent(s):
796a294
Upload 6 files
Browse files- config.json +25 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +13 -0
- tokenizer.json +0 -0
- tokenizer_config.json +10 -0
- trainer_state.json +154 -0
config.json
ADDED
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "EleutherAI/pythia-1.4b-deduped",
|
3 |
+
"architectures": [
|
4 |
+
"GPTNeoXForCausalLM"
|
5 |
+
],
|
6 |
+
"bos_token_id": 0,
|
7 |
+
"eos_token_id": 0,
|
8 |
+
"hidden_act": "gelu",
|
9 |
+
"hidden_size": 2048,
|
10 |
+
"initializer_range": 0.02,
|
11 |
+
"intermediate_size": 8192,
|
12 |
+
"layer_norm_eps": 1e-05,
|
13 |
+
"max_position_embeddings": 2048,
|
14 |
+
"model_type": "gpt_neox",
|
15 |
+
"num_attention_heads": 16,
|
16 |
+
"num_hidden_layers": 24,
|
17 |
+
"rotary_emb_base": 10000,
|
18 |
+
"rotary_pct": 0.25,
|
19 |
+
"tie_word_embeddings": false,
|
20 |
+
"torch_dtype": "float16",
|
21 |
+
"transformers_version": "4.25.1",
|
22 |
+
"use_cache": true,
|
23 |
+
"use_parallel_residual": true,
|
24 |
+
"vocab_size": 50281
|
25 |
+
}
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:9205c4e4dd515b5d129627740ded2dee5b3cc8e632dee4081317c408ad280a52
|
3 |
+
size 2929834572
|
special_tokens_map.json
ADDED
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"additional_special_tokens": [
|
3 |
+
"<prefix>",
|
4 |
+
"<bot>",
|
5 |
+
"<human>",
|
6 |
+
"</prefix>"
|
7 |
+
],
|
8 |
+
"bos_token": "<|endoftext|>",
|
9 |
+
"eos_token": "<|endoftext|>",
|
10 |
+
"pad_token": "<|padding|>",
|
11 |
+
"sep_token": "<|endoftext|>",
|
12 |
+
"unk_token": "<|endoftext|>"
|
13 |
+
}
|
tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
tokenizer_config.json
ADDED
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"add_prefix_space": false,
|
3 |
+
"bos_token": "<|endoftext|>",
|
4 |
+
"eos_token": "<|endoftext|>",
|
5 |
+
"model_max_length": 1000000000000000019884624838656,
|
6 |
+
"name_or_path": "EleutherAI/pythia-1.4b-deduped",
|
7 |
+
"special_tokens_map_file": "/fsx/home-hailey/.cache/huggingface/hub/models--EleutherAI--gpt-neox-20b/snapshots/3523781c8df75f7741687a4284f6f70e1afa12f4/special_tokens_map.json",
|
8 |
+
"tokenizer_class": "GPTNeoXTokenizer",
|
9 |
+
"unk_token": "<|endoftext|>"
|
10 |
+
}
|
trainer_state.json
ADDED
@@ -0,0 +1,154 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 7.675,
|
5 |
+
"global_step": 200,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.38,
|
12 |
+
"learning_rate": 4.999999999999999e-07,
|
13 |
+
"loss": 2.4017,
|
14 |
+
"step": 10
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.75,
|
18 |
+
"learning_rate": 6.505149978319905e-07,
|
19 |
+
"loss": 2.0997,
|
20 |
+
"step": 20
|
21 |
+
},
|
22 |
+
{
|
23 |
+
"epoch": 1.15,
|
24 |
+
"learning_rate": 7.38560627359831e-07,
|
25 |
+
"loss": 2.0895,
|
26 |
+
"step": 30
|
27 |
+
},
|
28 |
+
{
|
29 |
+
"epoch": 1.52,
|
30 |
+
"learning_rate": 8.01029995663981e-07,
|
31 |
+
"loss": 1.8155,
|
32 |
+
"step": 40
|
33 |
+
},
|
34 |
+
{
|
35 |
+
"epoch": 1.9,
|
36 |
+
"learning_rate": 8.494850021680092e-07,
|
37 |
+
"loss": 1.754,
|
38 |
+
"step": 50
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 2.3,
|
42 |
+
"learning_rate": 8.926649175053833e-07,
|
43 |
+
"loss": 1.9317,
|
44 |
+
"step": 60
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 2.67,
|
48 |
+
"learning_rate": 9.256291743595375e-07,
|
49 |
+
"loss": 1.7777,
|
50 |
+
"step": 70
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 3.08,
|
54 |
+
"learning_rate": 9.569069261918583e-07,
|
55 |
+
"loss": 1.8519,
|
56 |
+
"step": 80
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 3.45,
|
60 |
+
"learning_rate": 9.818939136727774e-07,
|
61 |
+
"loss": 1.706,
|
62 |
+
"step": 90
|
63 |
+
},
|
64 |
+
{
|
65 |
+
"epoch": 3.83,
|
66 |
+
"learning_rate": 9.991666666666667e-07,
|
67 |
+
"loss": 1.7139,
|
68 |
+
"step": 100
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 3.83,
|
72 |
+
"eval_oa_private_accuracy": 0.535490605427975,
|
73 |
+
"eval_oa_private_loss": 2.12109375,
|
74 |
+
"eval_oa_private_runtime": 1.2492,
|
75 |
+
"eval_oa_private_samples_per_second": 13.608,
|
76 |
+
"eval_oa_private_steps_per_second": 0.8,
|
77 |
+
"step": 100
|
78 |
+
},
|
79 |
+
{
|
80 |
+
"epoch": 4.22,
|
81 |
+
"learning_rate": 9.908333333333333e-07,
|
82 |
+
"loss": 1.7749,
|
83 |
+
"step": 110
|
84 |
+
},
|
85 |
+
{
|
86 |
+
"epoch": 4.6,
|
87 |
+
"learning_rate": 9.825e-07,
|
88 |
+
"loss": 1.6555,
|
89 |
+
"step": 120
|
90 |
+
},
|
91 |
+
{
|
92 |
+
"epoch": 4.97,
|
93 |
+
"learning_rate": 9.741666666666665e-07,
|
94 |
+
"loss": 1.7188,
|
95 |
+
"step": 130
|
96 |
+
},
|
97 |
+
{
|
98 |
+
"epoch": 5.38,
|
99 |
+
"learning_rate": 9.649999999999999e-07,
|
100 |
+
"loss": 1.7656,
|
101 |
+
"step": 140
|
102 |
+
},
|
103 |
+
{
|
104 |
+
"epoch": 5.75,
|
105 |
+
"learning_rate": 9.566666666666667e-07,
|
106 |
+
"loss": 1.6343,
|
107 |
+
"step": 150
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 6.15,
|
111 |
+
"learning_rate": 9.474999999999999e-07,
|
112 |
+
"loss": 1.6978,
|
113 |
+
"step": 160
|
114 |
+
},
|
115 |
+
{
|
116 |
+
"epoch": 6.53,
|
117 |
+
"learning_rate": 9.391666666666666e-07,
|
118 |
+
"loss": 1.6376,
|
119 |
+
"step": 170
|
120 |
+
},
|
121 |
+
{
|
122 |
+
"epoch": 6.9,
|
123 |
+
"learning_rate": 9.308333333333332e-07,
|
124 |
+
"loss": 1.6455,
|
125 |
+
"step": 180
|
126 |
+
},
|
127 |
+
{
|
128 |
+
"epoch": 7.3,
|
129 |
+
"learning_rate": 9.225e-07,
|
130 |
+
"loss": 1.7388,
|
131 |
+
"step": 190
|
132 |
+
},
|
133 |
+
{
|
134 |
+
"epoch": 7.67,
|
135 |
+
"learning_rate": 9.141666666666667e-07,
|
136 |
+
"loss": 1.6082,
|
137 |
+
"step": 200
|
138 |
+
},
|
139 |
+
{
|
140 |
+
"epoch": 7.67,
|
141 |
+
"eval_oa_private_accuracy": 0.6682947547249947,
|
142 |
+
"eval_oa_private_loss": 1.4609375,
|
143 |
+
"eval_oa_private_runtime": 0.6912,
|
144 |
+
"eval_oa_private_samples_per_second": 24.594,
|
145 |
+
"eval_oa_private_steps_per_second": 1.447,
|
146 |
+
"step": 200
|
147 |
+
}
|
148 |
+
],
|
149 |
+
"max_steps": 1300,
|
150 |
+
"num_train_epochs": 50,
|
151 |
+
"total_flos": 4.539212744569651e+17,
|
152 |
+
"trial_name": null,
|
153 |
+
"trial_params": null
|
154 |
+
}
|