Upload 11 files
Browse files- config.json +26 -0
- generation_config.json +7 -0
- model.safetensors +3 -0
- optimizer.pt +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +1 -0
- spiece.model +3 -0
- tokenizer_config.json +1 -0
- trainer_state.json +280 -0
- training_args.bin +3 -0
config.json
ADDED
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "/home/patrick/hugging_face/t5/mt5-large",
|
3 |
+
"architectures": [
|
4 |
+
"MT5ForConditionalGeneration"
|
5 |
+
],
|
6 |
+
"d_ff": 2816,
|
7 |
+
"d_kv": 64,
|
8 |
+
"d_model": 1024,
|
9 |
+
"decoder_start_token_id": 0,
|
10 |
+
"dropout_rate": 0.1,
|
11 |
+
"eos_token_id": 1,
|
12 |
+
"feed_forward_proj": "gated-gelu",
|
13 |
+
"initializer_factor": 1.0,
|
14 |
+
"is_encoder_decoder": true,
|
15 |
+
"layer_norm_epsilon": 1e-06,
|
16 |
+
"model_type": "mt5",
|
17 |
+
"num_decoder_layers": 24,
|
18 |
+
"num_heads": 16,
|
19 |
+
"num_layers": 24,
|
20 |
+
"output_past": true,
|
21 |
+
"pad_token_id": 0,
|
22 |
+
"relative_attention_num_buckets": 32,
|
23 |
+
"tie_word_embeddings": false,
|
24 |
+
"tokenizer_class": "T5Tokenizer",
|
25 |
+
"vocab_size": 250112
|
26 |
+
}
|
generation_config.json
ADDED
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_from_model_config": true,
|
3 |
+
"decoder_start_token_id": 0,
|
4 |
+
"eos_token_id": 1,
|
5 |
+
"pad_token_id": 0,
|
6 |
+
"transformers_version": "4.40.2"
|
7 |
+
}
|
model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:76ff7cfb57efc875e27624000392814a16dab31dbb0429a193ece8167e41af4d
|
3 |
+
size 4918393736
|
optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:208109da017f8c1520ebf394ddc447b6f51b5305d691fd6f44a61d944cb896a1
|
3 |
+
size 9837122384
|
rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c521e9deb3664d0cd507f97c9cf749f79420f711bf092c1d58e3c467d3b6fb7b
|
3 |
+
size 13990
|
scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4b7f151c12d99847478a0194bda24e6f18590939dbf74d0bbc868de7ff0c88f5
|
3 |
+
size 1064
|
special_tokens_map.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{"eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>"}
|
spiece.model
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ef78f86560d809067d12bac6c09f19a462cb3af3f54d2b8acbba26e1433125d6
|
3 |
+
size 4309802
|
tokenizer_config.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{"eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>", "extra_ids": 0, "additional_special_tokens": null, "special_tokens_map_file": "/home/patrick/.cache/torch/transformers/685ac0ca8568ec593a48b61b0a3c272beee9bc194a3c7241d15dcadb5f875e53.f76030f3ec1b96a8199b2593390c610e76ca8028ef3d24680000619ffb646276", "tokenizer_file": null, "name_or_path": "google/mt5-small"}
|
trainer_state.json
ADDED
@@ -0,0 +1,280 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 9.893048128342246,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 18500,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.26737967914438504,
|
13 |
+
"grad_norm": 2.1504368782043457,
|
14 |
+
"learning_rate": 4.8663101604278076e-05,
|
15 |
+
"loss": 7.6189,
|
16 |
+
"step": 500
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"epoch": 0.5347593582887701,
|
20 |
+
"grad_norm": 0.910918653011322,
|
21 |
+
"learning_rate": 4.732620320855615e-05,
|
22 |
+
"loss": 0.7574,
|
23 |
+
"step": 1000
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.8021390374331551,
|
27 |
+
"grad_norm": 0.7778192162513733,
|
28 |
+
"learning_rate": 4.598930481283423e-05,
|
29 |
+
"loss": 0.681,
|
30 |
+
"step": 1500
|
31 |
+
},
|
32 |
+
{
|
33 |
+
"epoch": 1.0695187165775402,
|
34 |
+
"grad_norm": 0.9133714437484741,
|
35 |
+
"learning_rate": 4.4652406417112304e-05,
|
36 |
+
"loss": 0.6446,
|
37 |
+
"step": 2000
|
38 |
+
},
|
39 |
+
{
|
40 |
+
"epoch": 1.3368983957219251,
|
41 |
+
"grad_norm": 1.6034926176071167,
|
42 |
+
"learning_rate": 4.331550802139038e-05,
|
43 |
+
"loss": 0.5912,
|
44 |
+
"step": 2500
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 1.6042780748663101,
|
48 |
+
"grad_norm": 0.7972973585128784,
|
49 |
+
"learning_rate": 4.197860962566845e-05,
|
50 |
+
"loss": 0.5784,
|
51 |
+
"step": 3000
|
52 |
+
},
|
53 |
+
{
|
54 |
+
"epoch": 1.8716577540106951,
|
55 |
+
"grad_norm": 1.0540673732757568,
|
56 |
+
"learning_rate": 4.0641711229946525e-05,
|
57 |
+
"loss": 0.5728,
|
58 |
+
"step": 3500
|
59 |
+
},
|
60 |
+
{
|
61 |
+
"epoch": 2.1390374331550803,
|
62 |
+
"grad_norm": 1.4703896045684814,
|
63 |
+
"learning_rate": 3.93048128342246e-05,
|
64 |
+
"loss": 0.5353,
|
65 |
+
"step": 4000
|
66 |
+
},
|
67 |
+
{
|
68 |
+
"epoch": 2.406417112299465,
|
69 |
+
"grad_norm": 1.052030086517334,
|
70 |
+
"learning_rate": 3.796791443850268e-05,
|
71 |
+
"loss": 0.5068,
|
72 |
+
"step": 4500
|
73 |
+
},
|
74 |
+
{
|
75 |
+
"epoch": 2.6737967914438503,
|
76 |
+
"grad_norm": 1.119246006011963,
|
77 |
+
"learning_rate": 3.6631016042780753e-05,
|
78 |
+
"loss": 0.5069,
|
79 |
+
"step": 5000
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 2.9411764705882355,
|
83 |
+
"grad_norm": 0.73586505651474,
|
84 |
+
"learning_rate": 3.529411764705883e-05,
|
85 |
+
"loss": 0.5034,
|
86 |
+
"step": 5500
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"epoch": 3.2085561497326203,
|
90 |
+
"grad_norm": 0.8226745128631592,
|
91 |
+
"learning_rate": 3.39572192513369e-05,
|
92 |
+
"loss": 0.4687,
|
93 |
+
"step": 6000
|
94 |
+
},
|
95 |
+
{
|
96 |
+
"epoch": 3.4759358288770055,
|
97 |
+
"grad_norm": 0.8904627561569214,
|
98 |
+
"learning_rate": 3.2620320855614975e-05,
|
99 |
+
"loss": 0.4591,
|
100 |
+
"step": 6500
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"epoch": 3.7433155080213902,
|
104 |
+
"grad_norm": 0.7578887939453125,
|
105 |
+
"learning_rate": 3.128342245989305e-05,
|
106 |
+
"loss": 0.4589,
|
107 |
+
"step": 7000
|
108 |
+
},
|
109 |
+
{
|
110 |
+
"epoch": 4.010695187165775,
|
111 |
+
"grad_norm": 0.8590931296348572,
|
112 |
+
"learning_rate": 2.9946524064171122e-05,
|
113 |
+
"loss": 0.4498,
|
114 |
+
"step": 7500
|
115 |
+
},
|
116 |
+
{
|
117 |
+
"epoch": 4.278074866310161,
|
118 |
+
"grad_norm": 0.8416089415550232,
|
119 |
+
"learning_rate": 2.8609625668449196e-05,
|
120 |
+
"loss": 0.4155,
|
121 |
+
"step": 8000
|
122 |
+
},
|
123 |
+
{
|
124 |
+
"epoch": 4.545454545454545,
|
125 |
+
"grad_norm": 0.7122125029563904,
|
126 |
+
"learning_rate": 2.7272727272727273e-05,
|
127 |
+
"loss": 0.4173,
|
128 |
+
"step": 8500
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 4.81283422459893,
|
132 |
+
"grad_norm": 0.7922715544700623,
|
133 |
+
"learning_rate": 2.5935828877005347e-05,
|
134 |
+
"loss": 0.4144,
|
135 |
+
"step": 9000
|
136 |
+
},
|
137 |
+
{
|
138 |
+
"epoch": 5.080213903743315,
|
139 |
+
"grad_norm": 1.4970874786376953,
|
140 |
+
"learning_rate": 2.4598930481283424e-05,
|
141 |
+
"loss": 0.4128,
|
142 |
+
"step": 9500
|
143 |
+
},
|
144 |
+
{
|
145 |
+
"epoch": 5.347593582887701,
|
146 |
+
"grad_norm": 0.9522221088409424,
|
147 |
+
"learning_rate": 2.32620320855615e-05,
|
148 |
+
"loss": 0.3862,
|
149 |
+
"step": 10000
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 5.614973262032086,
|
153 |
+
"grad_norm": 0.758745551109314,
|
154 |
+
"learning_rate": 2.192513368983957e-05,
|
155 |
+
"loss": 0.3889,
|
156 |
+
"step": 10500
|
157 |
+
},
|
158 |
+
{
|
159 |
+
"epoch": 5.882352941176471,
|
160 |
+
"grad_norm": 0.761408805847168,
|
161 |
+
"learning_rate": 2.058823529411765e-05,
|
162 |
+
"loss": 0.3787,
|
163 |
+
"step": 11000
|
164 |
+
},
|
165 |
+
{
|
166 |
+
"epoch": 6.149732620320855,
|
167 |
+
"grad_norm": 0.9881584644317627,
|
168 |
+
"learning_rate": 1.9251336898395722e-05,
|
169 |
+
"loss": 0.3762,
|
170 |
+
"step": 11500
|
171 |
+
},
|
172 |
+
{
|
173 |
+
"epoch": 6.4171122994652405,
|
174 |
+
"grad_norm": 0.6850549578666687,
|
175 |
+
"learning_rate": 1.7914438502673796e-05,
|
176 |
+
"loss": 0.3551,
|
177 |
+
"step": 12000
|
178 |
+
},
|
179 |
+
{
|
180 |
+
"epoch": 6.684491978609626,
|
181 |
+
"grad_norm": 1.1036587953567505,
|
182 |
+
"learning_rate": 1.6577540106951873e-05,
|
183 |
+
"loss": 0.3639,
|
184 |
+
"step": 12500
|
185 |
+
},
|
186 |
+
{
|
187 |
+
"epoch": 6.951871657754011,
|
188 |
+
"grad_norm": 1.33849036693573,
|
189 |
+
"learning_rate": 1.5240641711229947e-05,
|
190 |
+
"loss": 0.3615,
|
191 |
+
"step": 13000
|
192 |
+
},
|
193 |
+
{
|
194 |
+
"epoch": 7.219251336898395,
|
195 |
+
"grad_norm": 1.690130352973938,
|
196 |
+
"learning_rate": 1.3903743315508022e-05,
|
197 |
+
"loss": 0.3397,
|
198 |
+
"step": 13500
|
199 |
+
},
|
200 |
+
{
|
201 |
+
"epoch": 7.4866310160427805,
|
202 |
+
"grad_norm": 1.0579711198806763,
|
203 |
+
"learning_rate": 1.2566844919786098e-05,
|
204 |
+
"loss": 0.3454,
|
205 |
+
"step": 14000
|
206 |
+
},
|
207 |
+
{
|
208 |
+
"epoch": 7.754010695187166,
|
209 |
+
"grad_norm": 1.0972784757614136,
|
210 |
+
"learning_rate": 1.1229946524064172e-05,
|
211 |
+
"loss": 0.341,
|
212 |
+
"step": 14500
|
213 |
+
},
|
214 |
+
{
|
215 |
+
"epoch": 8.02139037433155,
|
216 |
+
"grad_norm": 0.9292057752609253,
|
217 |
+
"learning_rate": 9.893048128342247e-06,
|
218 |
+
"loss": 0.3458,
|
219 |
+
"step": 15000
|
220 |
+
},
|
221 |
+
{
|
222 |
+
"epoch": 8.288770053475936,
|
223 |
+
"grad_norm": 0.9827554821968079,
|
224 |
+
"learning_rate": 8.556149732620321e-06,
|
225 |
+
"loss": 0.3244,
|
226 |
+
"step": 15500
|
227 |
+
},
|
228 |
+
{
|
229 |
+
"epoch": 8.556149732620321,
|
230 |
+
"grad_norm": 1.4099150896072388,
|
231 |
+
"learning_rate": 7.2192513368983955e-06,
|
232 |
+
"loss": 0.3274,
|
233 |
+
"step": 16000
|
234 |
+
},
|
235 |
+
{
|
236 |
+
"epoch": 8.823529411764707,
|
237 |
+
"grad_norm": 1.0601311922073364,
|
238 |
+
"learning_rate": 5.882352941176471e-06,
|
239 |
+
"loss": 0.3267,
|
240 |
+
"step": 16500
|
241 |
+
},
|
242 |
+
{
|
243 |
+
"epoch": 9.090909090909092,
|
244 |
+
"grad_norm": 0.8129891157150269,
|
245 |
+
"learning_rate": 4.5454545454545455e-06,
|
246 |
+
"loss": 0.325,
|
247 |
+
"step": 17000
|
248 |
+
},
|
249 |
+
{
|
250 |
+
"epoch": 9.358288770053475,
|
251 |
+
"grad_norm": 0.862271249294281,
|
252 |
+
"learning_rate": 3.208556149732621e-06,
|
253 |
+
"loss": 0.3234,
|
254 |
+
"step": 17500
|
255 |
+
},
|
256 |
+
{
|
257 |
+
"epoch": 9.62566844919786,
|
258 |
+
"grad_norm": 0.8069100379943848,
|
259 |
+
"learning_rate": 1.8716577540106951e-06,
|
260 |
+
"loss": 0.3126,
|
261 |
+
"step": 18000
|
262 |
+
},
|
263 |
+
{
|
264 |
+
"epoch": 9.893048128342246,
|
265 |
+
"grad_norm": 0.9965262413024902,
|
266 |
+
"learning_rate": 5.347593582887701e-07,
|
267 |
+
"loss": 0.318,
|
268 |
+
"step": 18500
|
269 |
+
}
|
270 |
+
],
|
271 |
+
"logging_steps": 500,
|
272 |
+
"max_steps": 18700,
|
273 |
+
"num_input_tokens_seen": 0,
|
274 |
+
"num_train_epochs": 10,
|
275 |
+
"save_steps": 500,
|
276 |
+
"total_flos": 1.7282440998912e+16,
|
277 |
+
"train_batch_size": 8,
|
278 |
+
"trial_name": null,
|
279 |
+
"trial_params": null
|
280 |
+
}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e9e2436ae0d1940c88324306ad3eccf2d5afa5a3371ceae038915339b8d626fc
|
3 |
+
size 4920
|