andrew-healey commited on
Commit
637c5dd
·
verified ·
1 Parent(s): 3aa284e

Upload folder using huggingface_hub

Browse files
attention_kindself_n_heads4_seed1341/args.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"hellaswag": true, "attention_kind": "selective", "log_dir": "wider_is_better_4/attention_kindself_n_heads4_seed1341", "resume_checkpoint": null, "resume_optimizer": false, "add_a_head": false, "add_head_to_start": true, "new_head_init": "normal", "n_heads": 4, "protect_bos_token": true, "prevent_from_masking_myself": true, "max_steps": 10000, "warmup_steps": 200, "group": "wider_is_better_4", "use_wandb": true, "kill_self_after_run": false, "random_seed": 1341, "memory_penalty_epsilon": 0.1, "selection_head_linear_combo": "none", "selection_head_linear_combo_scale": 1.0, "protection_kind": "none", "leaky_relu_alpha": null, "leaky_relu_bias": null, "use_compile": true, "use_mini_model": false, "upload_to_hf": true, "seq_len": 256, "batch_size": 120, "total_batch_size": 30720, "protection_head_scaling_factor": 1.0, "protection_head_bias": 0.0, "n_sliced_masks": null, "n_latent_masks": null, "mask_layernorm": false, "residual_attention_masks": false, "compute_base_shapes": false, "base_shapes_savefile": null, "mup": true, "disable_selection": true, "mup_enable_coord_check_logging": false, "max_lr": 0.00015, "decay_lr": true, "readout_zero_init": false, "query_zero_init": false, "l1_loss": false, "debugpy": false, "key": "1.5e-4_30720_4_1341", "n_embd": 256}
attention_kindself_n_heads4_seed1341/dataloader_02500.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:db41c5e5513137877487a93451adf8ec4ed2448ab6e9471ebd5595c8e3293875
3
+ size 964
attention_kindself_n_heads4_seed1341/dataloader_05000.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f6df8947c6ab773db1947914387d3db345a84828521d3a64bae9b652e1b0a410
3
+ size 964
attention_kindself_n_heads4_seed1341/dataloader_07500.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:169891a726a7ff746d1a7aa99f459a66d85ceb4e9f2583f790f5b8501f97b6af
3
+ size 964
attention_kindself_n_heads4_seed1341/dataloader_09999.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e685a568a36c792ccbe7b5fcae0b9d630955e589991190bd8902836cea6a91df
3
+ size 964
attention_kindself_n_heads4_seed1341/log2.txt ADDED
@@ -0,0 +1,303 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ max_steps: 10000
2
+ 0 val loss 11.3001
3
+ 0 val perplexity 80832.1328
4
+ 0 train 11.306698 (lr=7.5000e-07) (hash(x)=145079536)
5
+ 100 val loss 9.3955
6
+ 100 val perplexity 12033.8779
7
+ 100 train 9.328828 (lr=7.5750e-05) (hash(x)=154745873)
8
+ 200 val loss 7.8945
9
+ 200 val perplexity 2682.5410
10
+ 200 train 8.046517 (lr=1.5000e-04) (hash(x)=155800595)
11
+ 300 val loss 7.5646
12
+ 300 val perplexity 1928.6428
13
+ 300 train 7.518878 (lr=1.4997e-04) (hash(x)=148595389)
14
+ 400 val loss 7.4628
15
+ 400 val perplexity 1741.9973
16
+ 400 train 7.364924 (lr=1.4986e-04) (hash(x)=145606733)
17
+ 500 val loss 7.3461
18
+ 500 val perplexity 1550.1819
19
+ 500 train 7.149421 (lr=1.4969e-04) (hash(x)=138221231)
20
+ 600 val loss 7.2483
21
+ 600 val perplexity 1405.6442
22
+ 600 train 7.251230 (lr=1.4945e-04) (hash(x)=150367139)
23
+ 700 val loss 7.1820
24
+ 700 val perplexity 1315.5784
25
+ 700 train 7.341902 (lr=1.4913e-04) (hash(x)=155579314)
26
+ 800 val loss 7.1344
27
+ 800 val perplexity 1254.3445
28
+ 800 train 7.203811 (lr=1.4876e-04) (hash(x)=155747374)
29
+ 900 val loss 7.0829
30
+ 900 val perplexity 1191.3647
31
+ 900 train 7.300850 (lr=1.4831e-04) (hash(x)=159334575)
32
+ 1000 val loss 7.0469
33
+ 1000 val perplexity 1149.2343
34
+ 1000 train 6.926498 (lr=1.4779e-04) (hash(x)=140604760)
35
+ 1100 val loss 7.0195
36
+ 1100 val perplexity 1118.2042
37
+ 1100 train 6.859845 (lr=1.4721e-04) (hash(x)=146208052)
38
+ 1200 val loss 6.9986
39
+ 1200 val perplexity 1095.0535
40
+ 1200 train 6.893322 (lr=1.4656e-04) (hash(x)=148404734)
41
+ 1300 val loss 6.9836
42
+ 1300 val perplexity 1078.7654
43
+ 1300 train 6.927601 (lr=1.4585e-04) (hash(x)=155681970)
44
+ 1400 val loss 6.9791
45
+ 1400 val perplexity 1073.9409
46
+ 1400 train 6.923635 (lr=1.4507e-04) (hash(x)=148115934)
47
+ 1500 val loss 6.9565
48
+ 1500 val perplexity 1049.9689
49
+ 1500 train 7.091454 (lr=1.4422e-04) (hash(x)=157074034)
50
+ 1600 val loss 6.9471
51
+ 1600 val perplexity 1040.1470
52
+ 1600 train 6.729518 (lr=1.4332e-04) (hash(x)=137464699)
53
+ 1700 val loss 6.9357
54
+ 1700 val perplexity 1028.3800
55
+ 1700 train 7.057977 (lr=1.4235e-04) (hash(x)=166955614)
56
+ 1800 val loss 6.9376
57
+ 1800 val perplexity 1030.3394
58
+ 1800 train 6.764245 (lr=1.4131e-04) (hash(x)=143886042)
59
+ 1900 val loss 6.9336
60
+ 1900 val perplexity 1026.2212
61
+ 1900 train 7.465428 (lr=1.4022e-04) (hash(x)=193610391)
62
+ 2000 val loss 6.9376
63
+ 2000 val perplexity 1030.2677
64
+ 2000 train 7.123946 (lr=1.3907e-04) (hash(x)=163799796)
65
+ 2100 val loss 6.9394
66
+ 2100 val perplexity 1032.1489
67
+ 2100 train 6.917353 (lr=1.3786e-04) (hash(x)=154107339)
68
+ 2200 val loss 6.9471
69
+ 2200 val perplexity 1040.1465
70
+ 2200 train 6.929567 (lr=1.3660e-04) (hash(x)=144592844)
71
+ 2300 val loss 6.9317
72
+ 2300 val perplexity 1024.1855
73
+ 2300 train 7.003430 (lr=1.3527e-04) (hash(x)=154535861)
74
+ 2400 val loss 6.9294
75
+ 2400 val perplexity 1021.9094
76
+ 2400 train 7.594524 (lr=1.3390e-04) (hash(x)=204706354)
77
+ 2500 val loss 6.9464
78
+ 2500 val perplexity 1039.4346
79
+ 2500 train 6.789571 (lr=1.3247e-04) (hash(x)=143783202)
80
+ 2600 val loss 6.9908
81
+ 2600 val perplexity 1086.5525
82
+ 2600 train 6.898761 (lr=1.3099e-04) (hash(x)=150862210)
83
+ 2700 val loss 7.0879
84
+ 2700 val perplexity 1197.4342
85
+ 2700 train 7.115728 (lr=1.2946e-04) (hash(x)=150753426)
86
+ 2800 val loss 7.0544
87
+ 2800 val perplexity 1157.9237
88
+ 2800 train 7.033858 (lr=1.2788e-04) (hash(x)=147766811)
89
+ 2900 val loss 7.0444
90
+ 2900 val perplexity 1146.4587
91
+ 2900 train 6.995884 (lr=1.2626e-04) (hash(x)=151177814)
92
+ 3000 val loss 7.0758
93
+ 3000 val perplexity 1182.9889
94
+ 3000 train 6.961751 (lr=1.2459e-04) (hash(x)=135925327)
95
+ 3100 val loss 7.0452
96
+ 3100 val perplexity 1147.3901
97
+ 3100 train 6.727612 (lr=1.2287e-04) (hash(x)=124842353)
98
+ 3200 val loss 7.0620
99
+ 3200 val perplexity 1166.8268
100
+ 3200 train 7.186998 (lr=1.2112e-04) (hash(x)=160440642)
101
+ 3300 val loss 7.0383
102
+ 3300 val perplexity 1139.4587
103
+ 3300 train 6.945041 (lr=1.1932e-04) (hash(x)=143017131)
104
+ 3400 val loss 7.0683
105
+ 3400 val perplexity 1174.1327
106
+ 3400 train 7.301206 (lr=1.1749e-04) (hash(x)=148143474)
107
+ 3500 val loss 7.0685
108
+ 3500 val perplexity 1174.3734
109
+ 3500 train 6.894324 (lr=1.1562e-04) (hash(x)=148732639)
110
+ 3600 val loss 7.0478
111
+ 3600 val perplexity 1150.2919
112
+ 3600 train 7.020499 (lr=1.1372e-04) (hash(x)=146333803)
113
+ 3700 val loss 7.0632
114
+ 3700 val perplexity 1168.2002
115
+ 3700 train 6.920694 (lr=1.1179e-04) (hash(x)=143611445)
116
+ 3800 val loss 7.0754
117
+ 3800 val perplexity 1182.5382
118
+ 3800 train 6.906082 (lr=1.0982e-04) (hash(x)=141272115)
119
+ 3900 val loss 7.0808
120
+ 3900 val perplexity 1188.9431
121
+ 3900 train 7.021241 (lr=1.0783e-04) (hash(x)=149161848)
122
+ 4000 val loss 7.0733
123
+ 4000 val perplexity 1180.0232
124
+ 4000 train 7.144450 (lr=1.0581e-04) (hash(x)=152036955)
125
+ 4100 val loss 7.0870
126
+ 4100 val perplexity 1196.3624
127
+ 4100 train 7.139550 (lr=1.0377e-04) (hash(x)=153260968)
128
+ 4200 val loss 7.0995
129
+ 4200 val perplexity 1211.3430
130
+ 4200 train 7.001007 (lr=1.0171e-04) (hash(x)=147629478)
131
+ 4300 val loss 7.1032
132
+ 4300 val perplexity 1215.8655
133
+ 4300 train 7.152019 (lr=9.9622e-05) (hash(x)=161701080)
134
+ 4400 val loss 7.1204
135
+ 4400 val perplexity 1236.9551
136
+ 4400 train 7.195014 (lr=9.7520e-05) (hash(x)=155290524)
137
+ 4500 val loss 7.1122
138
+ 4500 val perplexity 1226.8528
139
+ 4500 train 6.941273 (lr=9.5403e-05) (hash(x)=144726593)
140
+ 4600 val loss 7.0994
141
+ 4600 val perplexity 1211.2708
142
+ 4600 train 7.047345 (lr=9.3273e-05) (hash(x)=149517202)
143
+ 4700 val loss 7.1178
144
+ 4700 val perplexity 1233.7041
145
+ 4700 train 7.075057 (lr=9.1132e-05) (hash(x)=139403008)
146
+ 4800 val loss 7.1471
147
+ 4800 val perplexity 1270.3921
148
+ 4800 train 7.197305 (lr=8.8982e-05) (hash(x)=153387238)
149
+ 4900 val loss 7.1528
150
+ 4900 val perplexity 1277.6281
151
+ 4900 train 6.937476 (lr=8.6825e-05) (hash(x)=139881569)
152
+ 5000 val loss 7.1791
153
+ 5000 val perplexity 1311.7185
154
+ 5000 train 7.149471 (lr=8.4663e-05) (hash(x)=152602580)
155
+ 5100 val loss 7.1850
156
+ 5100 val perplexity 1319.4634
157
+ 5100 train 7.225554 (lr=8.2500e-05) (hash(x)=150598460)
158
+ 5200 val loss 7.1622
159
+ 5200 val perplexity 1289.7594
160
+ 5200 train 7.097391 (lr=8.0337e-05) (hash(x)=147660496)
161
+ 5300 val loss 7.2029
162
+ 5300 val perplexity 1343.2982
163
+ 5300 train 7.020668 (lr=7.8175e-05) (hash(x)=147693848)
164
+ 5400 val loss 7.1715
165
+ 5400 val perplexity 1301.7905
166
+ 5400 train 6.968065 (lr=7.6018e-05) (hash(x)=136048145)
167
+ 5500 val loss 7.1732
168
+ 5500 val perplexity 1304.0588
169
+ 5500 train 7.087708 (lr=7.3868e-05) (hash(x)=140313123)
170
+ 5600 val loss 7.1686
171
+ 5600 val perplexity 1298.0120
172
+ 5600 train 7.184072 (lr=7.1727e-05) (hash(x)=156997993)
173
+ 5700 val loss 7.1618
174
+ 5700 val perplexity 1289.2294
175
+ 5700 train 7.135228 (lr=6.9597e-05) (hash(x)=143513880)
176
+ 5800 val loss 7.1865
177
+ 5800 val perplexity 1321.4089
178
+ 5800 train 7.319064 (lr=6.7480e-05) (hash(x)=153736628)
179
+ 5900 val loss 7.1542
180
+ 5900 val perplexity 1279.5192
181
+ 5900 train 7.347223 (lr=6.5378e-05) (hash(x)=166385417)
182
+ 6000 val loss 7.1738
183
+ 6000 val perplexity 1304.7692
184
+ 6000 train 7.386747 (lr=6.3294e-05) (hash(x)=153216517)
185
+ 6100 val loss 7.1957
186
+ 6100 val perplexity 1333.7188
187
+ 6100 train 7.121440 (lr=6.1230e-05) (hash(x)=145387545)
188
+ 6200 val loss 7.1819
189
+ 6200 val perplexity 1315.4391
190
+ 6200 train 7.279177 (lr=5.9188e-05) (hash(x)=156306460)
191
+ 6300 val loss 7.1514
192
+ 6300 val perplexity 1275.8540
193
+ 6300 train 7.050121 (lr=5.7169e-05) (hash(x)=148610447)
194
+ 6400 val loss 7.1558
195
+ 6400 val perplexity 1281.5262
196
+ 6400 train 7.306859 (lr=5.5177e-05) (hash(x)=158343535)
197
+ 6500 val loss 7.1446
198
+ 6500 val perplexity 1267.2605
199
+ 6500 train 7.276059 (lr=5.3213e-05) (hash(x)=157742557)
200
+ 6600 val loss 7.1413
201
+ 6600 val perplexity 1263.0376
202
+ 6600 train 7.115061 (lr=5.1279e-05) (hash(x)=150457308)
203
+ 6700 val loss 7.1304
204
+ 6700 val perplexity 1249.4382
205
+ 6700 train 7.057846 (lr=4.9377e-05) (hash(x)=142426205)
206
+ 6800 val loss 7.1562
207
+ 6800 val perplexity 1282.0842
208
+ 6800 train 7.020480 (lr=4.7509e-05) (hash(x)=151522525)
209
+ 6900 val loss 7.1463
210
+ 6900 val perplexity 1269.3373
211
+ 6900 train 7.045939 (lr=4.5676e-05) (hash(x)=152029630)
212
+ 7000 val loss 7.1282
213
+ 7000 val perplexity 1246.6816
214
+ 7000 train 7.010334 (lr=4.3882e-05) (hash(x)=142978954)
215
+ 7100 val loss 7.1316
216
+ 7100 val perplexity 1250.8677
217
+ 7100 train 7.048053 (lr=4.2128e-05) (hash(x)=148064359)
218
+ 7200 val loss 7.1285
219
+ 7200 val perplexity 1247.0454
220
+ 7200 train 7.076699 (lr=4.0414e-05) (hash(x)=155771845)
221
+ 7300 val loss 7.1311
222
+ 7300 val perplexity 1250.2863
223
+ 7300 train 6.963120 (lr=3.8745e-05) (hash(x)=141639844)
224
+ 7400 val loss 7.1365
225
+ 7400 val perplexity 1257.0677
226
+ 7400 train 7.203446 (lr=3.7120e-05) (hash(x)=153466021)
227
+ 7500 val loss 7.1197
228
+ 7500 val perplexity 1236.1337
229
+ 7500 train 7.136865 (lr=3.5541e-05) (hash(x)=148843587)
230
+ 7600 val loss 7.1116
231
+ 7600 val perplexity 1226.1165
232
+ 7600 train 7.277462 (lr=3.4011e-05) (hash(x)=152053933)
233
+ 7700 val loss 7.1060
234
+ 7700 val perplexity 1219.2164
235
+ 7700 train 6.897977 (lr=3.2531e-05) (hash(x)=141564003)
236
+ 7800 val loss 7.1045
237
+ 7800 val perplexity 1217.3755
238
+ 7800 train 6.994552 (lr=3.1102e-05) (hash(x)=148617843)
239
+ 7900 val loss 7.1060
240
+ 7900 val perplexity 1219.3043
241
+ 7900 train 7.082467 (lr=2.9726e-05) (hash(x)=146417668)
242
+ 8000 val loss 7.1018
243
+ 8000 val perplexity 1214.1541
244
+ 8000 train 6.974138 (lr=2.8405e-05) (hash(x)=150165488)
245
+ 8100 val loss 7.0963
246
+ 8100 val perplexity 1207.5344
247
+ 8100 train 7.060221 (lr=2.7138e-05) (hash(x)=144818855)
248
+ 8200 val loss 7.0909
249
+ 8200 val perplexity 1201.0133
250
+ 8200 train 7.119874 (lr=2.5929e-05) (hash(x)=143074954)
251
+ 8300 val loss 7.0967
252
+ 8300 val perplexity 1208.0228
253
+ 8300 train 7.036266 (lr=2.4778e-05) (hash(x)=145172494)
254
+ 8400 val loss 7.0913
255
+ 8400 val perplexity 1201.4298
256
+ 8400 train 7.098520 (lr=2.3686e-05) (hash(x)=152284393)
257
+ 8500 val loss 7.0969
258
+ 8500 val perplexity 1208.2601
259
+ 8500 train 6.983650 (lr=2.2655e-05) (hash(x)=147044705)
260
+ 8600 val loss 7.0955
261
+ 8600 val perplexity 1206.5703
262
+ 8600 train 7.039187 (lr=2.1685e-05) (hash(x)=147762608)
263
+ 8700 val loss 7.0910
264
+ 8700 val perplexity 1201.1680
265
+ 8700 train 7.076082 (lr=2.0777e-05) (hash(x)=141994567)
266
+ 8800 val loss 7.0858
267
+ 8800 val perplexity 1194.8208
268
+ 8800 train 7.110641 (lr=1.9933e-05) (hash(x)=152552896)
269
+ 8900 val loss 7.0899
270
+ 8900 val perplexity 1199.7546
271
+ 8900 train 6.966035 (lr=1.9153e-05) (hash(x)=139714454)
272
+ 9000 val loss 7.0826
273
+ 9000 val perplexity 1191.0267
274
+ 9000 train 7.120393 (lr=1.8439e-05) (hash(x)=152274419)
275
+ 9100 val loss 7.0831
276
+ 9100 val perplexity 1191.6499
277
+ 9100 train 7.232574 (lr=1.7790e-05) (hash(x)=160536885)
278
+ 9200 val loss 7.0775
279
+ 9200 val perplexity 1185.0089
280
+ 9200 train 6.920467 (lr=1.7208e-05) (hash(x)=138386899)
281
+ 9300 val loss 7.0761
282
+ 9300 val perplexity 1183.3505
283
+ 9300 train 7.090115 (lr=1.6692e-05) (hash(x)=149957457)
284
+ 9400 val loss 7.0705
285
+ 9400 val perplexity 1176.6985
286
+ 9400 train 7.159608 (lr=1.6245e-05) (hash(x)=161671548)
287
+ 9500 val loss 7.0700
288
+ 9500 val perplexity 1176.1343
289
+ 9500 train 7.194458 (lr=1.5865e-05) (hash(x)=171857400)
290
+ 9600 val loss 7.0682
291
+ 9600 val perplexity 1173.9771
292
+ 9600 train 6.827000 (lr=1.5554e-05) (hash(x)=135378471)
293
+ 9700 val loss 7.0748
294
+ 9700 val perplexity 1181.8060
295
+ 9700 train 7.002652 (lr=1.5312e-05) (hash(x)=149514591)
296
+ 9800 val loss 7.0707
297
+ 9800 val perplexity 1176.9982
298
+ 9800 train 7.059768 (lr=1.5139e-05) (hash(x)=152679072)
299
+ 9900 val loss 7.0689
300
+ 9900 val perplexity 1174.8854
301
+ 9900 train 6.991833 (lr=1.5035e-05) (hash(x)=150415193)
302
+ 9999 val loss 7.0655
303
+ 9999 val perplexity 1170.8290
attention_kindself_n_heads4_seed1341/model_02500.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8c33d853b824a618231b03a5099efc279b9d9e5b64406a41441614ca3495a496
3
+ size 92843394
attention_kindself_n_heads4_seed1341/model_05000.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a7dfce5e2e2490fbc54169a381133a6a7f50c5bd6315114f93ec0d34ab57035b
3
+ size 92843394
attention_kindself_n_heads4_seed1341/model_07500.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f022e1f318a9ff5f71f8bf03ca4a0be59c7e19aec7499228c3cb84c234b02fbe
3
+ size 92843394
attention_kindself_n_heads4_seed1341/model_09999.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:41a5c62783abbee1a7e70f418aafd4affde384eb53f7607404494b593f2a22c2
3
+ size 92843394
attention_kindself_n_heads4_seed1341/optimizer_02500.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:767fa84c5f70a8e53dbf1419e39982a08fbe6e6fc5ae1b5c775d0feb0f1e7d9a
3
+ size 179406214
attention_kindself_n_heads4_seed1341/optimizer_05000.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:42bd6d830de914e5aabdd77f01551e171e6027880554a54365e1ed34959782d1
3
+ size 179406214
attention_kindself_n_heads4_seed1341/optimizer_07500.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f52907cfa74b5974aa1b2136db3f96d064bc659e38dd36c475a0c72d2119681f
3
+ size 179406214
attention_kindself_n_heads4_seed1341/optimizer_09999.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e77d542f2c6b2c551056a5b7ea5c142d14614182cc2e0410bc1f0990e310db03
3
+ size 179406214