Hugging Face's logo

w-ahmad
/
STRESS-TEST2

Safetensors
Model card Files
xet
 Community
w-ahmad commited on
Commit
32f2c91
·
verified ·
1 Parent(s): bf377e2

Auto upload zain 2026-08-14T23:07:46.851605 (part 4)

Browse files
Files changed (26) hide show
  1. zain/Activation/out/glu-waleed10-21L_run/checkpoint-1000/config.json +36 -0
  2. zain/Activation/out/glu-waleed10-21L_run/checkpoint-1000/model.safetensors +3 -0
  3. zain/Activation/out/glu-waleed10-21L_run/checkpoint-1000/optimizer.pt +3 -0
  4. zain/Activation/out/glu-waleed10-21L_run/checkpoint-1000/rng_state.pth +3 -0
  5. zain/Activation/out/glu-waleed10-21L_run/checkpoint-1000/scheduler.pt +3 -0
  6. zain/Activation/out/glu-waleed10-21L_run/checkpoint-1000/tokenizer.json +0 -0
  7. zain/Activation/out/glu-waleed10-21L_run/checkpoint-1000/tokenizer_config.json +13 -0
  8. zain/Activation/out/glu-waleed10-21L_run/checkpoint-1000/trainer_state.json +392 -0
  9. zain/Activation/out/glu-waleed10-21L_run/checkpoint-1000/training_args.bin +3 -0
  10. zain/Activation/out/glu-waleed10-21L_run/checkpoint-200/config.json +36 -0
  11. zain/Activation/out/glu-waleed10-21L_run/checkpoint-200/model.safetensors +3 -0
  12. zain/Activation/out/glu-waleed10-21L_run/checkpoint-200/optimizer.pt +3 -0
  13. zain/Activation/out/glu-waleed10-21L_run/checkpoint-200/rng_state.pth +3 -0
  14. zain/Activation/out/glu-waleed10-21L_run/checkpoint-200/scheduler.pt +3 -0
  15. zain/Activation/out/glu-waleed10-21L_run/checkpoint-200/tokenizer.json +0 -0
  16. zain/Activation/out/glu-waleed10-21L_run/checkpoint-200/tokenizer_config.json +13 -0
  17. zain/Activation/out/glu-waleed10-21L_run/checkpoint-200/trainer_state.json +104 -0
  18. zain/Activation/out/glu-waleed10-21L_run/checkpoint-200/training_args.bin +3 -0
  19. zain/Activation/out/glu-waleed10-21L_run/checkpoint-300/config.json +36 -0
  20. zain/Activation/out/glu-waleed10-21L_run/checkpoint-300/model.safetensors +3 -0
  21. zain/Activation/out/glu-waleed10-21L_run/checkpoint-300/optimizer.pt +3 -0
  22. zain/Activation/out/glu-waleed10-21L_run/checkpoint-300/rng_state.pth +3 -0
  23. zain/Activation/out/glu-waleed10-21L_run/checkpoint-300/scheduler.pt +3 -0
  24. zain/Activation/out/glu-waleed10-21L_run/checkpoint-300/tokenizer.json +0 -0
  25. zain/Activation/out/glu-waleed10-21L_run/checkpoint-300/tokenizer_config.json +13 -0
  26. zain/Activation/out/glu-waleed10-21L_run/checkpoint-300/trainer_state.json +139 -0
zain/Activation/out/glu-waleed10-21L_run/checkpoint-1000/config.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation": "waleed10",
3
+ "architectures": [
4
+ "TinyLlamaForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 1,
9
+ "dtype": "bfloat16",
10
+ "eos_token_id": 2,
11
+ "head_dim": 32,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 128,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 256,
16
+ "max_position_embeddings": 512,
17
+ "mlp_bias": false,
18
+ "mlp_type": "glu",
19
+ "model_type": "tiny_llama",
20
+ "num_attention_heads": 4,
21
+ "num_hidden_layers": 21,
22
+ "num_key_value_heads": 4,
23
+ "pad_token_id": 0,
24
+ "pretraining_tp": 1,
25
+ "rms_norm_eps": 1e-06,
26
+ "rope_parameters": {
27
+ "rope_theta": 10000.0,
28
+ "rope_type": "default"
29
+ },
30
+ "tie_word_embeddings": true,
31
+ "tokenizer_name": "w-ahmad/tiny-stories-tokenizer",
32
+ "transformers_version": "5.16.0.dev0",
33
+ "use_cache": false,
34
+ "vocab_size": 4096,
35
+ "waleed_beta": 10.0
36
+ }
zain/Activation/out/glu-waleed10-21L_run/checkpoint-1000/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:097fe7a792fe761865be6d5ef2032a06ec9ae849dce121d7a250219f84fecf10
3
+ size 7961528
zain/Activation/out/glu-waleed10-21L_run/checkpoint-1000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b87d1f2b03e708a2aa12577827b3306ce43ef83a3bb7695c9af4ba2235e6f516
3
+ size 16041274
zain/Activation/out/glu-waleed10-21L_run/checkpoint-1000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2b66e3cc7c452b707ddac5caf0aa17618afb9bc1a0333600a22c4afb353f3165
3
+ size 14244
zain/Activation/out/glu-waleed10-21L_run/checkpoint-1000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0fc3ca8af69cc9d003a05a139e7997960ab393fe31ae1d898b29038573704b4e
3
+ size 1064
zain/Activation/out/glu-waleed10-21L_run/checkpoint-1000/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
zain/Activation/out/glu-waleed10-21L_run/checkpoint-1000/tokenizer_config.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "backend": "tokenizers",
4
+ "bos_token": "<|endoftext|>",
5
+ "eos_token": "<|endoftext|>",
6
+ "errors": "replace",
7
+ "is_local": false,
8
+ "local_files_only": false,
9
+ "model_max_length": 1024,
10
+ "pad_token": "<|endoftext|>",
11
+ "tokenizer_class": "GPT2Tokenizer",
12
+ "unk_token": "<|endoftext|>"
13
+ }
zain/Activation/out/glu-waleed10-21L_run/checkpoint-1000/trainer_state.json ADDED
@@ -0,0 +1,392 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 0.5392289026691831,
6
+ "eval_steps": 60000,
7
+ "global_step": 1000,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.010784578053383662,
14
+ "grad_norm": 4.0,
15
+ "learning_rate": 0.001,
16
+ "loss": 28.876385498046876,
17
+ "step": 20
18
+ },
19
+ {
20
+ "epoch": 0.021569156106767323,
21
+ "grad_norm": 1.3984375,
22
+ "learning_rate": 0.001,
23
+ "loss": 24.010443115234374,
24
+ "step": 40
25
+ },
26
+ {
27
+ "epoch": 0.032353734160150985,
28
+ "grad_norm": 1.453125,
29
+ "learning_rate": 0.001,
30
+ "loss": 22.35637664794922,
31
+ "step": 60
32
+ },
33
+ {
34
+ "epoch": 0.043138312213534646,
35
+ "grad_norm": 3.34375,
36
+ "learning_rate": 0.001,
37
+ "loss": 20.596783447265626,
38
+ "step": 80
39
+ },
40
+ {
41
+ "epoch": 0.05392289026691831,
42
+ "grad_norm": 1.9140625,
43
+ "learning_rate": 0.001,
44
+ "loss": 18.65278015136719,
45
+ "step": 100
46
+ },
47
+ {
48
+ "epoch": 0.06470746832030197,
49
+ "grad_norm": 3.109375,
50
+ "learning_rate": 0.001,
51
+ "loss": 17.364892578125,
52
+ "step": 120
53
+ },
54
+ {
55
+ "epoch": 0.07549204637368563,
56
+ "grad_norm": 1.6015625,
57
+ "learning_rate": 0.001,
58
+ "loss": 16.395790100097656,
59
+ "step": 140
60
+ },
61
+ {
62
+ "epoch": 0.08627662442706929,
63
+ "grad_norm": 1.5234375,
64
+ "learning_rate": 0.001,
65
+ "loss": 15.567823791503907,
66
+ "step": 160
67
+ },
68
+ {
69
+ "epoch": 0.09706120248045295,
70
+ "grad_norm": 1.703125,
71
+ "learning_rate": 0.001,
72
+ "loss": 14.909393310546875,
73
+ "step": 180
74
+ },
75
+ {
76
+ "epoch": 0.10784578053383662,
77
+ "grad_norm": 2.4375,
78
+ "learning_rate": 0.001,
79
+ "loss": 14.338616943359375,
80
+ "step": 200
81
+ },
82
+ {
83
+ "epoch": 0.11863035858722028,
84
+ "grad_norm": 3.0625,
85
+ "learning_rate": 0.001,
86
+ "loss": 13.839776611328125,
87
+ "step": 220
88
+ },
89
+ {
90
+ "epoch": 0.12941493664060394,
91
+ "grad_norm": 1.5703125,
92
+ "learning_rate": 0.001,
93
+ "loss": 13.418243408203125,
94
+ "step": 240
95
+ },
96
+ {
97
+ "epoch": 0.1401995146939876,
98
+ "grad_norm": 1.96875,
99
+ "learning_rate": 0.001,
100
+ "loss": 12.980410766601562,
101
+ "step": 260
102
+ },
103
+ {
104
+ "epoch": 0.15098409274737126,
105
+ "grad_norm": 2.578125,
106
+ "learning_rate": 0.001,
107
+ "loss": 12.627825164794922,
108
+ "step": 280
109
+ },
110
+ {
111
+ "epoch": 0.1617686708007549,
112
+ "grad_norm": 2.078125,
113
+ "learning_rate": 0.001,
114
+ "loss": 12.301615142822266,
115
+ "step": 300
116
+ },
117
+ {
118
+ "epoch": 0.17255324885413859,
119
+ "grad_norm": 2.09375,
120
+ "learning_rate": 0.001,
121
+ "loss": 11.97824478149414,
122
+ "step": 320
123
+ },
124
+ {
125
+ "epoch": 0.18333782690752223,
126
+ "grad_norm": 2.171875,
127
+ "learning_rate": 0.001,
128
+ "loss": 11.688941955566406,
129
+ "step": 340
130
+ },
131
+ {
132
+ "epoch": 0.1941224049609059,
133
+ "grad_norm": 2.125,
134
+ "learning_rate": 0.001,
135
+ "loss": 11.42403335571289,
136
+ "step": 360
137
+ },
138
+ {
139
+ "epoch": 0.20490698301428956,
140
+ "grad_norm": 2.234375,
141
+ "learning_rate": 0.001,
142
+ "loss": 11.158559417724609,
143
+ "step": 380
144
+ },
145
+ {
146
+ "epoch": 0.21569156106767323,
147
+ "grad_norm": 2.453125,
148
+ "learning_rate": 0.001,
149
+ "loss": 10.936893463134766,
150
+ "step": 400
151
+ },
152
+ {
153
+ "epoch": 0.22647613912105688,
154
+ "grad_norm": 2.078125,
155
+ "learning_rate": 0.001,
156
+ "loss": 10.734546661376953,
157
+ "step": 420
158
+ },
159
+ {
160
+ "epoch": 0.23726071717444056,
161
+ "grad_norm": 2.71875,
162
+ "learning_rate": 0.001,
163
+ "loss": 10.598049926757813,
164
+ "step": 440
165
+ },
166
+ {
167
+ "epoch": 0.2480452952278242,
168
+ "grad_norm": 2.515625,
169
+ "learning_rate": 0.001,
170
+ "loss": 10.398246002197265,
171
+ "step": 460
172
+ },
173
+ {
174
+ "epoch": 0.2588298732812079,
175
+ "grad_norm": 2.390625,
176
+ "learning_rate": 0.001,
177
+ "loss": 10.239346313476563,
178
+ "step": 480
179
+ },
180
+ {
181
+ "epoch": 0.26961445133459155,
182
+ "grad_norm": 2.25,
183
+ "learning_rate": 0.001,
184
+ "loss": 10.090744018554688,
185
+ "step": 500
186
+ },
187
+ {
188
+ "epoch": 0.2803990293879752,
189
+ "grad_norm": 2.21875,
190
+ "learning_rate": 0.001,
191
+ "loss": 9.968287658691406,
192
+ "step": 520
193
+ },
194
+ {
195
+ "epoch": 0.29118360744135885,
196
+ "grad_norm": 2.015625,
197
+ "learning_rate": 0.001,
198
+ "loss": 9.856298828125,
199
+ "step": 540
200
+ },
201
+ {
202
+ "epoch": 0.3019681854947425,
203
+ "grad_norm": 2.4375,
204
+ "learning_rate": 0.001,
205
+ "loss": 9.739572906494141,
206
+ "step": 560
207
+ },
208
+ {
209
+ "epoch": 0.3127527635481262,
210
+ "grad_norm": 1.9765625,
211
+ "learning_rate": 0.001,
212
+ "loss": 9.626911163330078,
213
+ "step": 580
214
+ },
215
+ {
216
+ "epoch": 0.3235373416015098,
217
+ "grad_norm": 2.109375,
218
+ "learning_rate": 0.001,
219
+ "loss": 9.514720153808593,
220
+ "step": 600
221
+ },
222
+ {
223
+ "epoch": 0.3343219196548935,
224
+ "grad_norm": 2.203125,
225
+ "learning_rate": 0.001,
226
+ "loss": 9.425536346435546,
227
+ "step": 620
228
+ },
229
+ {
230
+ "epoch": 0.34510649770827717,
231
+ "grad_norm": 2.296875,
232
+ "learning_rate": 0.001,
233
+ "loss": 9.341368865966796,
234
+ "step": 640
235
+ },
236
+ {
237
+ "epoch": 0.35589107576166085,
238
+ "grad_norm": 1.8203125,
239
+ "learning_rate": 0.001,
240
+ "loss": 9.254900360107422,
241
+ "step": 660
242
+ },
243
+ {
244
+ "epoch": 0.36667565381504447,
245
+ "grad_norm": 2.484375,
246
+ "learning_rate": 0.001,
247
+ "loss": 9.180448913574219,
248
+ "step": 680
249
+ },
250
+ {
251
+ "epoch": 0.37746023186842814,
252
+ "grad_norm": 2.515625,
253
+ "learning_rate": 0.001,
254
+ "loss": 9.107149505615235,
255
+ "step": 700
256
+ },
257
+ {
258
+ "epoch": 0.3882448099218118,
259
+ "grad_norm": 2.0,
260
+ "learning_rate": 0.001,
261
+ "loss": 9.010431671142578,
262
+ "step": 720
263
+ },
264
+ {
265
+ "epoch": 0.3990293879751955,
266
+ "grad_norm": 2.15625,
267
+ "learning_rate": 0.001,
268
+ "loss": 8.976707458496094,
269
+ "step": 740
270
+ },
271
+ {
272
+ "epoch": 0.4098139660285791,
273
+ "grad_norm": 1.9609375,
274
+ "learning_rate": 0.001,
275
+ "loss": 8.91417999267578,
276
+ "step": 760
277
+ },
278
+ {
279
+ "epoch": 0.4205985440819628,
280
+ "grad_norm": 1.8828125,
281
+ "learning_rate": 0.001,
282
+ "loss": 8.871702575683594,
283
+ "step": 780
284
+ },
285
+ {
286
+ "epoch": 0.43138312213534646,
287
+ "grad_norm": 2.0,
288
+ "learning_rate": 0.001,
289
+ "loss": 8.769727325439453,
290
+ "step": 800
291
+ },
292
+ {
293
+ "epoch": 0.44216770018873014,
294
+ "grad_norm": 2.0,
295
+ "learning_rate": 0.001,
296
+ "loss": 8.744908142089844,
297
+ "step": 820
298
+ },
299
+ {
300
+ "epoch": 0.45295227824211376,
301
+ "grad_norm": 2.0625,
302
+ "learning_rate": 0.001,
303
+ "loss": 8.702173614501953,
304
+ "step": 840
305
+ },
306
+ {
307
+ "epoch": 0.46373685629549743,
308
+ "grad_norm": 2.171875,
309
+ "learning_rate": 0.001,
310
+ "loss": 8.637160491943359,
311
+ "step": 860
312
+ },
313
+ {
314
+ "epoch": 0.4745214343488811,
315
+ "grad_norm": 2.078125,
316
+ "learning_rate": 0.001,
317
+ "loss": 8.599034118652344,
318
+ "step": 880
319
+ },
320
+ {
321
+ "epoch": 0.4853060124022648,
322
+ "grad_norm": 2.0625,
323
+ "learning_rate": 0.001,
324
+ "loss": 8.549813842773437,
325
+ "step": 900
326
+ },
327
+ {
328
+ "epoch": 0.4960905904556484,
329
+ "grad_norm": 2.015625,
330
+ "learning_rate": 0.001,
331
+ "loss": 8.507289123535156,
332
+ "step": 920
333
+ },
334
+ {
335
+ "epoch": 0.5068751685090321,
336
+ "grad_norm": 1.90625,
337
+ "learning_rate": 0.001,
338
+ "loss": 8.488274383544923,
339
+ "step": 940
340
+ },
341
+ {
342
+ "epoch": 0.5176597465624158,
343
+ "grad_norm": 1.8671875,
344
+ "learning_rate": 0.001,
345
+ "loss": 8.468278503417968,
346
+ "step": 960
347
+ },
348
+ {
349
+ "epoch": 0.5284443246157994,
350
+ "grad_norm": 1.671875,
351
+ "learning_rate": 0.001,
352
+ "loss": 8.434804534912109,
353
+ "step": 980
354
+ },
355
+ {
356
+ "epoch": 0.5392289026691831,
357
+ "grad_norm": 2.09375,
358
+ "learning_rate": 0.001,
359
+ "loss": 8.39569091796875,
360
+ "step": 1000
361
+ },
362
+ {
363
+ "epoch": 0.5392289026691831,
364
+ "eval_loss": 2.1011769771575928,
365
+ "eval_runtime": 9.4243,
366
+ "eval_samples_per_second": 1010.893,
367
+ "eval_steps_per_second": 7.958,
368
+ "step": 1000
369
+ }
370
+ ],
371
+ "logging_steps": 20,
372
+ "max_steps": 1000,
373
+ "num_input_tokens_seen": 0,
374
+ "num_train_epochs": 1,
375
+ "save_steps": 100,
376
+ "stateful_callbacks": {
377
+ "TrainerControl": {
378
+ "args": {
379
+ "should_epoch_stop": false,
380
+ "should_evaluate": false,
381
+ "should_log": false,
382
+ "should_save": true,
383
+ "should_training_stop": true
384
+ },
385
+ "attributes": {}
386
+ }
387
+ },
388
+ "total_flos": 5420315836416000.0,
389
+ "train_batch_size": 128,
390
+ "trial_name": null,
391
+ "trial_params": null
392
+ }
zain/Activation/out/glu-waleed10-21L_run/checkpoint-1000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d4a5fb74d92c2a824150af3ee3051d6e747acb606e30ff7dc799c234f6c87bd3
3
+ size 4920
zain/Activation/out/glu-waleed10-21L_run/checkpoint-200/config.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation": "waleed10",
3
+ "architectures": [
4
+ "TinyLlamaForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 1,
9
+ "dtype": "bfloat16",
10
+ "eos_token_id": 2,
11
+ "head_dim": 32,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 128,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 256,
16
+ "max_position_embeddings": 512,
17
+ "mlp_bias": false,
18
+ "mlp_type": "glu",
19
+ "model_type": "tiny_llama",
20
+ "num_attention_heads": 4,
21
+ "num_hidden_layers": 21,
22
+ "num_key_value_heads": 4,
23
+ "pad_token_id": 0,
24
+ "pretraining_tp": 1,
25
+ "rms_norm_eps": 1e-06,
26
+ "rope_parameters": {
27
+ "rope_theta": 10000.0,
28
+ "rope_type": "default"
29
+ },
30
+ "tie_word_embeddings": true,
31
+ "tokenizer_name": "w-ahmad/tiny-stories-tokenizer",
32
+ "transformers_version": "5.16.0.dev0",
33
+ "use_cache": false,
34
+ "vocab_size": 4096,
35
+ "waleed_beta": 10.0
36
+ }
zain/Activation/out/glu-waleed10-21L_run/checkpoint-200/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2f5e9a7cf4808fe26ee56fcf92545f220789ea11a8f2210aaa06efe5589fd32d
3
+ size 7961528
zain/Activation/out/glu-waleed10-21L_run/checkpoint-200/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a3e160ecba8a28924cd3c5d8036b2d3e19598e1e31605b8175da89b16165a336
3
+ size 16041274
zain/Activation/out/glu-waleed10-21L_run/checkpoint-200/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9d9cd6a0487226e5bd30d1846894c82af483733ab4381b75bae9c0745e05d405
3
+ size 14244
zain/Activation/out/glu-waleed10-21L_run/checkpoint-200/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2d3b5a970fc92af42664e84ba686059eaaacfaa668691c0cf6d2afddda677461
3
+ size 1064
zain/Activation/out/glu-waleed10-21L_run/checkpoint-200/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
zain/Activation/out/glu-waleed10-21L_run/checkpoint-200/tokenizer_config.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "backend": "tokenizers",
4
+ "bos_token": "<|endoftext|>",
5
+ "eos_token": "<|endoftext|>",
6
+ "errors": "replace",
7
+ "is_local": false,
8
+ "local_files_only": false,
9
+ "model_max_length": 1024,
10
+ "pad_token": "<|endoftext|>",
11
+ "tokenizer_class": "GPT2Tokenizer",
12
+ "unk_token": "<|endoftext|>"
13
+ }
zain/Activation/out/glu-waleed10-21L_run/checkpoint-200/trainer_state.json ADDED
@@ -0,0 +1,104 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 0.10784578053383662,
6
+ "eval_steps": 60000,
7
+ "global_step": 200,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.010784578053383662,
14
+ "grad_norm": 4.0,
15
+ "learning_rate": 0.001,
16
+ "loss": 28.876385498046876,
17
+ "step": 20
18
+ },
19
+ {
20
+ "epoch": 0.021569156106767323,
21
+ "grad_norm": 1.3984375,
22
+ "learning_rate": 0.001,
23
+ "loss": 24.010443115234374,
24
+ "step": 40
25
+ },
26
+ {
27
+ "epoch": 0.032353734160150985,
28
+ "grad_norm": 1.453125,
29
+ "learning_rate": 0.001,
30
+ "loss": 22.35637664794922,
31
+ "step": 60
32
+ },
33
+ {
34
+ "epoch": 0.043138312213534646,
35
+ "grad_norm": 3.34375,
36
+ "learning_rate": 0.001,
37
+ "loss": 20.596783447265626,
38
+ "step": 80
39
+ },
40
+ {
41
+ "epoch": 0.05392289026691831,
42
+ "grad_norm": 1.9140625,
43
+ "learning_rate": 0.001,
44
+ "loss": 18.65278015136719,
45
+ "step": 100
46
+ },
47
+ {
48
+ "epoch": 0.06470746832030197,
49
+ "grad_norm": 3.109375,
50
+ "learning_rate": 0.001,
51
+ "loss": 17.364892578125,
52
+ "step": 120
53
+ },
54
+ {
55
+ "epoch": 0.07549204637368563,
56
+ "grad_norm": 1.6015625,
57
+ "learning_rate": 0.001,
58
+ "loss": 16.395790100097656,
59
+ "step": 140
60
+ },
61
+ {
62
+ "epoch": 0.08627662442706929,
63
+ "grad_norm": 1.5234375,
64
+ "learning_rate": 0.001,
65
+ "loss": 15.567823791503907,
66
+ "step": 160
67
+ },
68
+ {
69
+ "epoch": 0.09706120248045295,
70
+ "grad_norm": 1.703125,
71
+ "learning_rate": 0.001,
72
+ "loss": 14.909393310546875,
73
+ "step": 180
74
+ },
75
+ {
76
+ "epoch": 0.10784578053383662,
77
+ "grad_norm": 2.4375,
78
+ "learning_rate": 0.001,
79
+ "loss": 14.338616943359375,
80
+ "step": 200
81
+ }
82
+ ],
83
+ "logging_steps": 20,
84
+ "max_steps": 1000,
85
+ "num_input_tokens_seen": 0,
86
+ "num_train_epochs": 1,
87
+ "save_steps": 100,
88
+ "stateful_callbacks": {
89
+ "TrainerControl": {
90
+ "args": {
91
+ "should_epoch_stop": false,
92
+ "should_evaluate": false,
93
+ "should_log": false,
94
+ "should_save": true,
95
+ "should_training_stop": false
96
+ },
97
+ "attributes": {}
98
+ }
99
+ },
100
+ "total_flos": 1084063167283200.0,
101
+ "train_batch_size": 128,
102
+ "trial_name": null,
103
+ "trial_params": null
104
+ }
zain/Activation/out/glu-waleed10-21L_run/checkpoint-200/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d4a5fb74d92c2a824150af3ee3051d6e747acb606e30ff7dc799c234f6c87bd3
3
+ size 4920
zain/Activation/out/glu-waleed10-21L_run/checkpoint-300/config.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation": "waleed10",
3
+ "architectures": [
4
+ "TinyLlamaForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 1,
9
+ "dtype": "bfloat16",
10
+ "eos_token_id": 2,
11
+ "head_dim": 32,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 128,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 256,
16
+ "max_position_embeddings": 512,
17
+ "mlp_bias": false,
18
+ "mlp_type": "glu",
19
+ "model_type": "tiny_llama",
20
+ "num_attention_heads": 4,
21
+ "num_hidden_layers": 21,
22
+ "num_key_value_heads": 4,
23
+ "pad_token_id": 0,
24
+ "pretraining_tp": 1,
25
+ "rms_norm_eps": 1e-06,
26
+ "rope_parameters": {
27
+ "rope_theta": 10000.0,
28
+ "rope_type": "default"
29
+ },
30
+ "tie_word_embeddings": true,
31
+ "tokenizer_name": "w-ahmad/tiny-stories-tokenizer",
32
+ "transformers_version": "5.16.0.dev0",
33
+ "use_cache": false,
34
+ "vocab_size": 4096,
35
+ "waleed_beta": 10.0
36
+ }
zain/Activation/out/glu-waleed10-21L_run/checkpoint-300/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c21faca07cd0a2a08958b65bb66f3c062b2442b289136b692959ebadac8fe5f8
3
+ size 7961528
zain/Activation/out/glu-waleed10-21L_run/checkpoint-300/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a9d4eb9bec546eaf5d9de7a9f89f1e2853868cf64a6c19247e1911250c569e60
3
+ size 16041274
zain/Activation/out/glu-waleed10-21L_run/checkpoint-300/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9d9cd6a0487226e5bd30d1846894c82af483733ab4381b75bae9c0745e05d405
3
+ size 14244
zain/Activation/out/glu-waleed10-21L_run/checkpoint-300/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:76cd21f32221b90fb8288fd2428595ec8b5039fe2425086d67cacb5c8d814c9e
3
+ size 1064
zain/Activation/out/glu-waleed10-21L_run/checkpoint-300/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
zain/Activation/out/glu-waleed10-21L_run/checkpoint-300/tokenizer_config.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "backend": "tokenizers",
4
+ "bos_token": "<|endoftext|>",
5
+ "eos_token": "<|endoftext|>",
6
+ "errors": "replace",
7
+ "is_local": false,
8
+ "local_files_only": false,
9
+ "model_max_length": 1024,
10
+ "pad_token": "<|endoftext|>",
11
+ "tokenizer_class": "GPT2Tokenizer",
12
+ "unk_token": "<|endoftext|>"
13
+ }
zain/Activation/out/glu-waleed10-21L_run/checkpoint-300/trainer_state.json ADDED
@@ -0,0 +1,139 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 0.1617686708007549,
6
+ "eval_steps": 60000,
7
+ "global_step": 300,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.010784578053383662,
14
+ "grad_norm": 4.0,
15
+ "learning_rate": 0.001,
16
+ "loss": 28.876385498046876,
17
+ "step": 20
18
+ },
19
+ {
20
+ "epoch": 0.021569156106767323,
21
+ "grad_norm": 1.3984375,
22
+ "learning_rate": 0.001,
23
+ "loss": 24.010443115234374,
24
+ "step": 40
25
+ },
26
+ {
27
+ "epoch": 0.032353734160150985,
28
+ "grad_norm": 1.453125,
29
+ "learning_rate": 0.001,
30
+ "loss": 22.35637664794922,
31
+ "step": 60
32
+ },
33
+ {
34
+ "epoch": 0.043138312213534646,
35
+ "grad_norm": 3.34375,
36
+ "learning_rate": 0.001,
37
+ "loss": 20.596783447265626,
38
+ "step": 80
39
+ },
40
+ {
41
+ "epoch": 0.05392289026691831,
42
+ "grad_norm": 1.9140625,
43
+ "learning_rate": 0.001,
44
+ "loss": 18.65278015136719,
45
+ "step": 100
46
+ },
47
+ {
48
+ "epoch": 0.06470746832030197,
49
+ "grad_norm": 3.109375,
50
+ "learning_rate": 0.001,
51
+ "loss": 17.364892578125,
52
+ "step": 120
53
+ },
54
+ {
55
+ "epoch": 0.07549204637368563,
56
+ "grad_norm": 1.6015625,
57
+ "learning_rate": 0.001,
58
+ "loss": 16.395790100097656,
59
+ "step": 140
60
+ },
61
+ {
62
+ "epoch": 0.08627662442706929,
63
+ "grad_norm": 1.5234375,
64
+ "learning_rate": 0.001,
65
+ "loss": 15.567823791503907,
66
+ "step": 160
67
+ },
68
+ {
69
+ "epoch": 0.09706120248045295,
70
+ "grad_norm": 1.703125,
71
+ "learning_rate": 0.001,
72
+ "loss": 14.909393310546875,
73
+ "step": 180
74
+ },
75
+ {
76
+ "epoch": 0.10784578053383662,
77
+ "grad_norm": 2.4375,
78
+ "learning_rate": 0.001,
79
+ "loss": 14.338616943359375,
80
+ "step": 200
81
+ },
82
+ {
83
+ "epoch": 0.11863035858722028,
84
+ "grad_norm": 3.0625,
85
+ "learning_rate": 0.001,
86
+ "loss": 13.839776611328125,
87
+ "step": 220
88
+ },
89
+ {
90
+ "epoch": 0.12941493664060394,
91
+ "grad_norm": 1.5703125,
92
+ "learning_rate": 0.001,
93
+ "loss": 13.418243408203125,
94
+ "step": 240
95
+ },
96
+ {
97
+ "epoch": 0.1401995146939876,
98
+ "grad_norm": 1.96875,
99
+ "learning_rate": 0.001,
100
+ "loss": 12.980410766601562,
101
+ "step": 260
102
+ },
103
+ {
104
+ "epoch": 0.15098409274737126,
105
+ "grad_norm": 2.578125,
106
+ "learning_rate": 0.001,
107
+ "loss": 12.627825164794922,
108
+ "step": 280
109
+ },
110
+ {
111
+ "epoch": 0.1617686708007549,
112
+ "grad_norm": 2.078125,
113
+ "learning_rate": 0.001,
114
+ "loss": 12.301615142822266,
115
+ "step": 300
116
+ }
117
+ ],
118
+ "logging_steps": 20,
119
+ "max_steps": 1000,
120
+ "num_input_tokens_seen": 0,
121
+ "num_train_epochs": 1,
122
+ "save_steps": 100,
123
+ "stateful_callbacks": {
124
+ "TrainerControl": {
125
+ "args": {
126
+ "should_epoch_stop": false,
127
+ "should_evaluate": false,
128
+ "should_log": false,
129
+ "should_save": true,
130
+ "should_training_stop": false
131
+ },
132
+ "attributes": {}
133
+ }
134
+ },
135
+ "total_flos": 1626094750924800.0,
136
+ "train_batch_size": 128,
137
+ "trial_name": null,
138
+ "trial_params": null
139
+ }