anrilombard
commited on
Commit
•
ea7feed
1
Parent(s):
589ff97
Training in progress, step 485000
Browse files- manual/checkpoint-17000/config.json +40 -0
- manual/checkpoint-17000/generation_config.json +7 -0
- manual/checkpoint-17000/model.safetensors +3 -0
- manual/checkpoint-17000/optimizer.pt +3 -0
- manual/checkpoint-17000/rng_state.pth +3 -0
- manual/checkpoint-17000/scheduler.pt +3 -0
- manual/checkpoint-17000/trainer_state.json +271 -0
- manual/checkpoint-17000/training_args.bin +3 -0
- model.safetensors +1 -1
- training_args.bin +1 -1
manual/checkpoint-17000/config.json
ADDED
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "trainer/configs/large_config.json",
|
3 |
+
"activation_function": "gelu_new",
|
4 |
+
"architectures": [
|
5 |
+
"SAFEDoubleHeadsModel"
|
6 |
+
],
|
7 |
+
"attn_pdrop": 0.1,
|
8 |
+
"bos_token_id": 1,
|
9 |
+
"embd_pdrop": 0.1,
|
10 |
+
"eos_token_id": 2,
|
11 |
+
"id2label": {
|
12 |
+
"0": "LABEL_0"
|
13 |
+
},
|
14 |
+
"initializer_range": 0.02,
|
15 |
+
"label2id": {
|
16 |
+
"LABEL_0": 0
|
17 |
+
},
|
18 |
+
"layer_norm_epsilon": 1e-05,
|
19 |
+
"model_type": "gpt2",
|
20 |
+
"n_embd": 660,
|
21 |
+
"n_head": 10,
|
22 |
+
"n_inner": null,
|
23 |
+
"n_layer": 10,
|
24 |
+
"n_positions": 1024,
|
25 |
+
"pad_token_id": 3,
|
26 |
+
"reorder_and_upcast_attn": false,
|
27 |
+
"resid_pdrop": 0.1,
|
28 |
+
"scale_attn_by_inverse_layer_idx": false,
|
29 |
+
"scale_attn_weights": true,
|
30 |
+
"summary_activation": "tanh",
|
31 |
+
"summary_first_dropout": 0.1,
|
32 |
+
"summary_hidden_size": 128,
|
33 |
+
"summary_proj_to_labels": true,
|
34 |
+
"summary_type": "cls_index",
|
35 |
+
"summary_use_proj": true,
|
36 |
+
"torch_dtype": "float32",
|
37 |
+
"transformers_version": "4.42.3",
|
38 |
+
"use_cache": true,
|
39 |
+
"vocab_size": 1880
|
40 |
+
}
|
manual/checkpoint-17000/generation_config.json
ADDED
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_from_model_config": true,
|
3 |
+
"bos_token_id": 1,
|
4 |
+
"eos_token_id": 2,
|
5 |
+
"pad_token_id": 3,
|
6 |
+
"transformers_version": "4.42.3"
|
7 |
+
}
|
manual/checkpoint-17000/model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b601a3ed10eef25f03ee3e8ab8ff8a6a9f7e8197f258d333f9726ea6f5ce4220
|
3 |
+
size 217454844
|
manual/checkpoint-17000/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:16186e34e3a93366787c45d0851d4c2e4154e00aed7823170bd2fc8bbf676cf6
|
3 |
+
size 434307002
|
manual/checkpoint-17000/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b1346e9edf4eb92556d7e6a530ba146df909f84d7279c1ecee8a0595a021fa75
|
3 |
+
size 14244
|
manual/checkpoint-17000/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:d7d9c20964d6b5816b3dbf35c5f6287479777a7ae3da85a79e7ca4f93bf07ccb
|
3 |
+
size 1064
|
manual/checkpoint-17000/trainer_state.json
ADDED
@@ -0,0 +1,271 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 0.17,
|
5 |
+
"eval_steps": 500.0,
|
6 |
+
"global_step": 17000,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.005,
|
13 |
+
"grad_norm": 2.999999761581421,
|
14 |
+
"learning_rate": 5.000000000000001e-07,
|
15 |
+
"loss": 4.7795,
|
16 |
+
"step": 500
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.01,
|
20 |
+
"grad_norm": 2.9999992847442627,
|
21 |
+
"learning_rate": 1.0000000000000002e-06,
|
22 |
+
"loss": 2.4619,
|
23 |
+
"step": 1000
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.015,
|
27 |
+
"grad_norm": 2.999999761581421,
|
28 |
+
"learning_rate": 1.5e-06,
|
29 |
+
"loss": 2.0368,
|
30 |
+
"step": 1500
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 0.02,
|
34 |
+
"grad_norm": 2.9999992847442627,
|
35 |
+
"learning_rate": 2.0000000000000003e-06,
|
36 |
+
"loss": 1.7481,
|
37 |
+
"step": 2000
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 0.025,
|
41 |
+
"grad_norm": 2.999999761581421,
|
42 |
+
"learning_rate": 2.5e-06,
|
43 |
+
"loss": 1.4947,
|
44 |
+
"step": 2500
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.03,
|
48 |
+
"grad_norm": 2.9999992847442627,
|
49 |
+
"learning_rate": 3e-06,
|
50 |
+
"loss": 1.3007,
|
51 |
+
"step": 3000
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 0.035,
|
55 |
+
"grad_norm": 2.9999990463256836,
|
56 |
+
"learning_rate": 3.5e-06,
|
57 |
+
"loss": 1.1635,
|
58 |
+
"step": 3500
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 0.04,
|
62 |
+
"grad_norm": 2.999999761581421,
|
63 |
+
"learning_rate": 4.000000000000001e-06,
|
64 |
+
"loss": 1.0721,
|
65 |
+
"step": 4000
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 0.045,
|
69 |
+
"grad_norm": 2.9999990463256836,
|
70 |
+
"learning_rate": 4.5e-06,
|
71 |
+
"loss": 1.0072,
|
72 |
+
"step": 4500
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 0.05,
|
76 |
+
"grad_norm": 2.9999992847442627,
|
77 |
+
"learning_rate": 5e-06,
|
78 |
+
"loss": 0.9584,
|
79 |
+
"step": 5000
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 0.055,
|
83 |
+
"grad_norm": 2.9999992847442627,
|
84 |
+
"learning_rate": 5.500000000000001e-06,
|
85 |
+
"loss": 0.918,
|
86 |
+
"step": 5500
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 0.06,
|
90 |
+
"grad_norm": 2.9999992847442627,
|
91 |
+
"learning_rate": 6e-06,
|
92 |
+
"loss": 0.8862,
|
93 |
+
"step": 6000
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 0.065,
|
97 |
+
"grad_norm": 2.9999992847442627,
|
98 |
+
"learning_rate": 6.5000000000000004e-06,
|
99 |
+
"loss": 0.8607,
|
100 |
+
"step": 6500
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 0.07,
|
104 |
+
"grad_norm": 2.9999990463256836,
|
105 |
+
"learning_rate": 7e-06,
|
106 |
+
"loss": 0.8363,
|
107 |
+
"step": 7000
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 0.075,
|
111 |
+
"grad_norm": 2.9999992847442627,
|
112 |
+
"learning_rate": 7.500000000000001e-06,
|
113 |
+
"loss": 0.8174,
|
114 |
+
"step": 7500
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 0.08,
|
118 |
+
"grad_norm": 2.9999992847442627,
|
119 |
+
"learning_rate": 8.000000000000001e-06,
|
120 |
+
"loss": 0.8015,
|
121 |
+
"step": 8000
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 0.085,
|
125 |
+
"grad_norm": 2.999999523162842,
|
126 |
+
"learning_rate": 8.5e-06,
|
127 |
+
"loss": 0.7885,
|
128 |
+
"step": 8500
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 0.09,
|
132 |
+
"grad_norm": 2.762766122817993,
|
133 |
+
"learning_rate": 9e-06,
|
134 |
+
"loss": 0.7744,
|
135 |
+
"step": 9000
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 0.095,
|
139 |
+
"grad_norm": 2.5979771614074707,
|
140 |
+
"learning_rate": 9.5e-06,
|
141 |
+
"loss": 0.765,
|
142 |
+
"step": 9500
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 0.1,
|
146 |
+
"grad_norm": 2.4132823944091797,
|
147 |
+
"learning_rate": 1e-05,
|
148 |
+
"loss": 0.7546,
|
149 |
+
"step": 10000
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 0.105,
|
153 |
+
"grad_norm": 2.9999990463256836,
|
154 |
+
"learning_rate": 9.944444444444445e-06,
|
155 |
+
"loss": 0.7454,
|
156 |
+
"step": 10500
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 0.11,
|
160 |
+
"grad_norm": 1.9988462924957275,
|
161 |
+
"learning_rate": 9.88888888888889e-06,
|
162 |
+
"loss": 0.7354,
|
163 |
+
"step": 11000
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 0.115,
|
167 |
+
"grad_norm": 2.1960322856903076,
|
168 |
+
"learning_rate": 9.833333333333333e-06,
|
169 |
+
"loss": 0.7274,
|
170 |
+
"step": 11500
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 0.12,
|
174 |
+
"grad_norm": 2.3398022651672363,
|
175 |
+
"learning_rate": 9.777777777777779e-06,
|
176 |
+
"loss": 0.7211,
|
177 |
+
"step": 12000
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 0.125,
|
181 |
+
"grad_norm": 1.904077172279358,
|
182 |
+
"learning_rate": 9.722222222222223e-06,
|
183 |
+
"loss": 0.7152,
|
184 |
+
"step": 12500
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 0.13,
|
188 |
+
"grad_norm": 2.774881601333618,
|
189 |
+
"learning_rate": 9.666666666666667e-06,
|
190 |
+
"loss": 0.7099,
|
191 |
+
"step": 13000
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 0.135,
|
195 |
+
"grad_norm": 1.7885903120040894,
|
196 |
+
"learning_rate": 9.611111111111112e-06,
|
197 |
+
"loss": 0.7046,
|
198 |
+
"step": 13500
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 0.14,
|
202 |
+
"grad_norm": 1.9303686618804932,
|
203 |
+
"learning_rate": 9.555555555555556e-06,
|
204 |
+
"loss": 0.6997,
|
205 |
+
"step": 14000
|
206 |
+
},
|
207 |
+
{
|
208 |
+
"epoch": 0.145,
|
209 |
+
"grad_norm": 2.0811753273010254,
|
210 |
+
"learning_rate": 9.5e-06,
|
211 |
+
"loss": 0.6949,
|
212 |
+
"step": 14500
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 0.15,
|
216 |
+
"grad_norm": 2.0698301792144775,
|
217 |
+
"learning_rate": 9.444444444444445e-06,
|
218 |
+
"loss": 0.6918,
|
219 |
+
"step": 15000
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 0.155,
|
223 |
+
"grad_norm": 1.838094711303711,
|
224 |
+
"learning_rate": 9.38888888888889e-06,
|
225 |
+
"loss": 0.688,
|
226 |
+
"step": 15500
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 0.16,
|
230 |
+
"grad_norm": 1.7373666763305664,
|
231 |
+
"learning_rate": 9.333333333333334e-06,
|
232 |
+
"loss": 0.6835,
|
233 |
+
"step": 16000
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 0.165,
|
237 |
+
"grad_norm": 2.1607325077056885,
|
238 |
+
"learning_rate": 9.277777777777778e-06,
|
239 |
+
"loss": 0.6802,
|
240 |
+
"step": 16500
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 0.17,
|
244 |
+
"grad_norm": 1.8940073251724243,
|
245 |
+
"learning_rate": 9.222222222222224e-06,
|
246 |
+
"loss": 0.6761,
|
247 |
+
"step": 17000
|
248 |
+
}
|
249 |
+
],
|
250 |
+
"logging_steps": 500,
|
251 |
+
"max_steps": 100000,
|
252 |
+
"num_input_tokens_seen": 0,
|
253 |
+
"num_train_epochs": 9223372036854775807,
|
254 |
+
"save_steps": 1000,
|
255 |
+
"stateful_callbacks": {
|
256 |
+
"TrainerControl": {
|
257 |
+
"args": {
|
258 |
+
"should_epoch_stop": false,
|
259 |
+
"should_evaluate": false,
|
260 |
+
"should_log": false,
|
261 |
+
"should_save": true,
|
262 |
+
"should_training_stop": false
|
263 |
+
},
|
264 |
+
"attributes": {}
|
265 |
+
}
|
266 |
+
},
|
267 |
+
"total_flos": 4.273433586508646e+16,
|
268 |
+
"train_batch_size": 64,
|
269 |
+
"trial_name": null,
|
270 |
+
"trial_params": null
|
271 |
+
}
|
manual/checkpoint-17000/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e59b3a3018b4bac8c30c130dc256bae308fe3db1ac00a19ed74630792abf832e
|
3 |
+
size 5304
|
model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 217454844
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8a98bf99e9eb418ca24f3398cc6ccf53c4083fb63f4c4820cab8d843feed8c30
|
3 |
size 217454844
|
training_args.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 5304
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c4b950106c8844bc1e46b02a9a7e75d3c88cba83db826edf2d777fae7a97508b
|
3 |
size 5304
|