w-ahmad commited on
Commit
33cb782
·
verified ·
1 Parent(s): 46b4b52

Auto upload zain 2026-08-14T20:28:11.532417

Browse files
zain/Activation/out/glu-silu-waleed10-9L_run/checkpoint-1000/config.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation": "silu-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": 9,
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-silu-waleed10-9L_run/checkpoint-1000/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ca7a808ba9383f2a1f20a31108f60a1dd105c32aefaf496bec84c001f010d893
3
+ size 4011496
zain/Activation/out/glu-silu-waleed10-9L_run/checkpoint-1000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7624c7cda3121d310ba26376368abfeb723e898514313a35761e67796f7a4812
3
+ size 8074746
zain/Activation/out/glu-silu-waleed10-9L_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-silu-waleed10-9L_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-silu-waleed10-9L_run/checkpoint-1000/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
zain/Activation/out/glu-silu-waleed10-9L_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-silu-waleed10-9L_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": 0.5390835579514824,
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": 0.01078167115902965,
14
+ "grad_norm": 0.97265625,
15
+ "learning_rate": 0.001,
16
+ "loss": 7.1676788330078125,
17
+ "step": 20
18
+ },
19
+ {
20
+ "epoch": 0.0215633423180593,
21
+ "grad_norm": 0.609375,
22
+ "learning_rate": 0.001,
23
+ "loss": 5.993187332153321,
24
+ "step": 40
25
+ },
26
+ {
27
+ "epoch": 0.03234501347708895,
28
+ "grad_norm": 0.48828125,
29
+ "learning_rate": 0.001,
30
+ "loss": 5.496040725708008,
31
+ "step": 60
32
+ },
33
+ {
34
+ "epoch": 0.0431266846361186,
35
+ "grad_norm": 0.54296875,
36
+ "learning_rate": 0.001,
37
+ "loss": 4.934689331054687,
38
+ "step": 80
39
+ },
40
+ {
41
+ "epoch": 0.05390835579514825,
42
+ "grad_norm": 0.41796875,
43
+ "learning_rate": 0.001,
44
+ "loss": 4.479576110839844,
45
+ "step": 100
46
+ },
47
+ {
48
+ "epoch": 0.0646900269541779,
49
+ "grad_norm": 0.37890625,
50
+ "learning_rate": 0.001,
51
+ "loss": 4.171505355834961,
52
+ "step": 120
53
+ },
54
+ {
55
+ "epoch": 0.07547169811320754,
56
+ "grad_norm": 0.55078125,
57
+ "learning_rate": 0.001,
58
+ "loss": 3.940719223022461,
59
+ "step": 140
60
+ },
61
+ {
62
+ "epoch": 0.0862533692722372,
63
+ "grad_norm": 0.80859375,
64
+ "learning_rate": 0.001,
65
+ "loss": 3.752634811401367,
66
+ "step": 160
67
+ },
68
+ {
69
+ "epoch": 0.09703504043126684,
70
+ "grad_norm": 0.4765625,
71
+ "learning_rate": 0.001,
72
+ "loss": 3.6034828186035157,
73
+ "step": 180
74
+ },
75
+ {
76
+ "epoch": 0.1078167115902965,
77
+ "grad_norm": 0.59375,
78
+ "learning_rate": 0.001,
79
+ "loss": 3.4787067413330077,
80
+ "step": 200
81
+ },
82
+ {
83
+ "epoch": 0.11859838274932614,
84
+ "grad_norm": 0.9296875,
85
+ "learning_rate": 0.001,
86
+ "loss": 3.3636749267578123,
87
+ "step": 220
88
+ },
89
+ {
90
+ "epoch": 0.1293800539083558,
91
+ "grad_norm": 0.703125,
92
+ "learning_rate": 0.001,
93
+ "loss": 3.269498825073242,
94
+ "step": 240
95
+ },
96
+ {
97
+ "epoch": 0.14016172506738545,
98
+ "grad_norm": 0.50390625,
99
+ "learning_rate": 0.001,
100
+ "loss": 3.183236503601074,
101
+ "step": 260
102
+ },
103
+ {
104
+ "epoch": 0.1509433962264151,
105
+ "grad_norm": 0.6171875,
106
+ "learning_rate": 0.001,
107
+ "loss": 3.099096488952637,
108
+ "step": 280
109
+ },
110
+ {
111
+ "epoch": 0.16172506738544473,
112
+ "grad_norm": 0.8359375,
113
+ "learning_rate": 0.001,
114
+ "loss": 3.0244571685791017,
115
+ "step": 300
116
+ },
117
+ {
118
+ "epoch": 0.1725067385444744,
119
+ "grad_norm": 0.51953125,
120
+ "learning_rate": 0.001,
121
+ "loss": 2.9644407272338866,
122
+ "step": 320
123
+ },
124
+ {
125
+ "epoch": 0.18328840970350405,
126
+ "grad_norm": 0.83984375,
127
+ "learning_rate": 0.001,
128
+ "loss": 2.8989635467529298,
129
+ "step": 340
130
+ },
131
+ {
132
+ "epoch": 0.1940700808625337,
133
+ "grad_norm": 0.443359375,
134
+ "learning_rate": 0.001,
135
+ "loss": 2.851881980895996,
136
+ "step": 360
137
+ },
138
+ {
139
+ "epoch": 0.20485175202156333,
140
+ "grad_norm": 0.72265625,
141
+ "learning_rate": 0.001,
142
+ "loss": 2.790276527404785,
143
+ "step": 380
144
+ },
145
+ {
146
+ "epoch": 0.215633423180593,
147
+ "grad_norm": 0.80078125,
148
+ "learning_rate": 0.001,
149
+ "loss": 2.7615821838378904,
150
+ "step": 400
151
+ },
152
+ {
153
+ "epoch": 0.22641509433962265,
154
+ "grad_norm": 0.83203125,
155
+ "learning_rate": 0.001,
156
+ "loss": 2.7132925033569335,
157
+ "step": 420
158
+ },
159
+ {
160
+ "epoch": 0.2371967654986523,
161
+ "grad_norm": 0.66796875,
162
+ "learning_rate": 0.001,
163
+ "loss": 2.691997528076172,
164
+ "step": 440
165
+ },
166
+ {
167
+ "epoch": 0.24797843665768193,
168
+ "grad_norm": 0.69921875,
169
+ "learning_rate": 0.001,
170
+ "loss": 2.6550079345703126,
171
+ "step": 460
172
+ },
173
+ {
174
+ "epoch": 0.2587601078167116,
175
+ "grad_norm": 0.58984375,
176
+ "learning_rate": 0.001,
177
+ "loss": 2.6208978652954102,
178
+ "step": 480
179
+ },
180
+ {
181
+ "epoch": 0.2695417789757412,
182
+ "grad_norm": 0.8046875,
183
+ "learning_rate": 0.001,
184
+ "loss": 2.593034362792969,
185
+ "step": 500
186
+ },
187
+ {
188
+ "epoch": 0.2803234501347709,
189
+ "grad_norm": 0.87890625,
190
+ "learning_rate": 0.001,
191
+ "loss": 2.5671127319335936,
192
+ "step": 520
193
+ },
194
+ {
195
+ "epoch": 0.29110512129380056,
196
+ "grad_norm": 0.6875,
197
+ "learning_rate": 0.001,
198
+ "loss": 2.5521316528320312,
199
+ "step": 540
200
+ },
201
+ {
202
+ "epoch": 0.3018867924528302,
203
+ "grad_norm": 0.59375,
204
+ "learning_rate": 0.001,
205
+ "loss": 2.5245645523071287,
206
+ "step": 560
207
+ },
208
+ {
209
+ "epoch": 0.31266846361185985,
210
+ "grad_norm": 0.58203125,
211
+ "learning_rate": 0.001,
212
+ "loss": 2.5037845611572265,
213
+ "step": 580
214
+ },
215
+ {
216
+ "epoch": 0.32345013477088946,
217
+ "grad_norm": 0.74609375,
218
+ "learning_rate": 0.001,
219
+ "loss": 2.482448196411133,
220
+ "step": 600
221
+ },
222
+ {
223
+ "epoch": 0.33423180592991913,
224
+ "grad_norm": 1.078125,
225
+ "learning_rate": 0.001,
226
+ "loss": 2.4656560897827147,
227
+ "step": 620
228
+ },
229
+ {
230
+ "epoch": 0.3450134770889488,
231
+ "grad_norm": 0.5703125,
232
+ "learning_rate": 0.001,
233
+ "loss": 2.4476362228393556,
234
+ "step": 640
235
+ },
236
+ {
237
+ "epoch": 0.3557951482479784,
238
+ "grad_norm": 0.6953125,
239
+ "learning_rate": 0.001,
240
+ "loss": 2.4330963134765624,
241
+ "step": 660
242
+ },
243
+ {
244
+ "epoch": 0.3665768194070081,
245
+ "grad_norm": 0.66796875,
246
+ "learning_rate": 0.001,
247
+ "loss": 2.41370849609375,
248
+ "step": 680
249
+ },
250
+ {
251
+ "epoch": 0.37735849056603776,
252
+ "grad_norm": 0.80078125,
253
+ "learning_rate": 0.001,
254
+ "loss": 2.397682952880859,
255
+ "step": 700
256
+ },
257
+ {
258
+ "epoch": 0.3881401617250674,
259
+ "grad_norm": 0.625,
260
+ "learning_rate": 0.001,
261
+ "loss": 2.3781137466430664,
262
+ "step": 720
263
+ },
264
+ {
265
+ "epoch": 0.39892183288409705,
266
+ "grad_norm": 0.64453125,
267
+ "learning_rate": 0.001,
268
+ "loss": 2.3710350036621093,
269
+ "step": 740
270
+ },
271
+ {
272
+ "epoch": 0.40970350404312667,
273
+ "grad_norm": 0.7265625,
274
+ "learning_rate": 0.001,
275
+ "loss": 2.3579540252685547,
276
+ "step": 760
277
+ },
278
+ {
279
+ "epoch": 0.42048517520215634,
280
+ "grad_norm": 0.6484375,
281
+ "learning_rate": 0.001,
282
+ "loss": 2.34680233001709,
283
+ "step": 780
284
+ },
285
+ {
286
+ "epoch": 0.431266846361186,
287
+ "grad_norm": 0.84375,
288
+ "learning_rate": 0.001,
289
+ "loss": 2.325694274902344,
290
+ "step": 800
291
+ },
292
+ {
293
+ "epoch": 0.4420485175202156,
294
+ "grad_norm": 0.62109375,
295
+ "learning_rate": 0.001,
296
+ "loss": 2.32082576751709,
297
+ "step": 820
298
+ },
299
+ {
300
+ "epoch": 0.4528301886792453,
301
+ "grad_norm": 0.78515625,
302
+ "learning_rate": 0.001,
303
+ "loss": 2.3087087631225587,
304
+ "step": 840
305
+ },
306
+ {
307
+ "epoch": 0.4636118598382749,
308
+ "grad_norm": 0.59765625,
309
+ "learning_rate": 0.001,
310
+ "loss": 2.293583869934082,
311
+ "step": 860
312
+ },
313
+ {
314
+ "epoch": 0.4743935309973046,
315
+ "grad_norm": 0.78125,
316
+ "learning_rate": 0.001,
317
+ "loss": 2.283832550048828,
318
+ "step": 880
319
+ },
320
+ {
321
+ "epoch": 0.48517520215633425,
322
+ "grad_norm": 0.4765625,
323
+ "learning_rate": 0.001,
324
+ "loss": 2.272108268737793,
325
+ "step": 900
326
+ },
327
+ {
328
+ "epoch": 0.49595687331536387,
329
+ "grad_norm": 0.64453125,
330
+ "learning_rate": 0.001,
331
+ "loss": 2.263412284851074,
332
+ "step": 920
333
+ },
334
+ {
335
+ "epoch": 0.5067385444743935,
336
+ "grad_norm": 0.63671875,
337
+ "learning_rate": 0.001,
338
+ "loss": 2.261708450317383,
339
+ "step": 940
340
+ },
341
+ {
342
+ "epoch": 0.5175202156334232,
343
+ "grad_norm": 0.6484375,
344
+ "learning_rate": 0.001,
345
+ "loss": 2.255706214904785,
346
+ "step": 960
347
+ },
348
+ {
349
+ "epoch": 0.5283018867924528,
350
+ "grad_norm": 0.6015625,
351
+ "learning_rate": 0.001,
352
+ "loss": 2.24764461517334,
353
+ "step": 980
354
+ },
355
+ {
356
+ "epoch": 0.5390835579514824,
357
+ "grad_norm": 0.90234375,
358
+ "learning_rate": 0.001,
359
+ "loss": 2.2394777297973634,
360
+ "step": 1000
361
+ },
362
+ {
363
+ "epoch": 0.5390835579514824,
364
+ "eval_loss": 2.241689443588257,
365
+ "eval_runtime": 9.1259,
366
+ "eval_samples_per_second": 1043.957,
367
+ "eval_steps_per_second": 2.082,
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": 2323107545088000.0,
389
+ "train_batch_size": 512,
390
+ "trial_name": null,
391
+ "trial_params": null
392
+ }
zain/Activation/out/glu-silu-waleed10-9L_run/checkpoint-1000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ea57eafa5d8d7c92fa44d42011d21fc8a9b019b0c1ee506d1d2f3d7c5eb1dccd
3
+ size 4920