w-ahmad commited on
Commit
ce93686
·
verified ·
1 Parent(s): 2aa7ba0

Auto upload zain 2026-08-12T22:55:26.034780

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. zain/Activation/out/glu-situglu-100L_run/checkpoint-1800/config.json +36 -0
  2. zain/Activation/out/glu-situglu-100L_run/checkpoint-1800/model.safetensors +3 -0
  3. zain/Activation/out/glu-situglu-100L_run/checkpoint-1800/optimizer.pt +3 -0
  4. zain/Activation/out/glu-situglu-100L_run/checkpoint-1800/rng_state.pth +3 -0
  5. zain/Activation/out/glu-situglu-100L_run/checkpoint-1800/scheduler.pt +3 -0
  6. zain/Activation/out/glu-situglu-100L_run/checkpoint-1800/tokenizer.json +0 -0
  7. zain/Activation/out/glu-situglu-100L_run/checkpoint-1800/tokenizer_config.json +13 -0
  8. zain/Activation/out/glu-situglu-100L_run/checkpoint-1800/trainer_state.json +664 -0
  9. zain/Activation/out/glu-situglu-100L_run/checkpoint-1800/training_args.bin +3 -0
  10. zain/Activation/out/glu-situglu-100L_run/checkpoint-1900/config.json +36 -0
  11. zain/Activation/out/glu-situglu-100L_run/checkpoint-1900/model.safetensors +3 -0
  12. zain/Activation/out/glu-situglu-100L_run/checkpoint-1900/optimizer.pt +3 -0
  13. zain/Activation/out/glu-situglu-100L_run/checkpoint-1900/rng_state.pth +3 -0
  14. zain/Activation/out/glu-situglu-100L_run/checkpoint-1900/scheduler.pt +3 -0
  15. zain/Activation/out/glu-situglu-100L_run/checkpoint-1900/tokenizer.json +0 -0
  16. zain/Activation/out/glu-situglu-100L_run/checkpoint-1900/tokenizer_config.json +13 -0
  17. zain/Activation/out/glu-situglu-100L_run/checkpoint-1900/trainer_state.json +699 -0
  18. zain/Activation/out/glu-situglu-100L_run/checkpoint-1900/training_args.bin +3 -0
  19. zain/Activation/out/glu-situglu-100L_run/checkpoint-2000/config.json +36 -0
  20. zain/Activation/out/glu-situglu-100L_run/checkpoint-2000/model.safetensors +3 -0
  21. zain/Activation/out/glu-situglu-100L_run/checkpoint-2000/optimizer.pt +3 -0
  22. zain/Activation/out/glu-situglu-100L_run/checkpoint-2000/rng_state.pth +3 -0
  23. zain/Activation/out/glu-situglu-100L_run/checkpoint-2000/scheduler.pt +3 -0
  24. zain/Activation/out/glu-situglu-100L_run/checkpoint-2000/tokenizer.json +0 -0
  25. zain/Activation/out/glu-situglu-100L_run/checkpoint-2000/tokenizer_config.json +13 -0
  26. zain/Activation/out/glu-situglu-100L_run/checkpoint-2000/trainer_state.json +734 -0
  27. zain/Activation/out/glu-situglu-100L_run/checkpoint-2000/training_args.bin +3 -0
  28. zain/Activation/out/glu-situglu-100L_run/checkpoint-2100/config.json +36 -0
  29. zain/Activation/out/glu-situglu-100L_run/checkpoint-2100/model.safetensors +3 -0
  30. zain/Activation/out/glu-situglu-100L_run/checkpoint-2100/optimizer.pt +3 -0
  31. zain/Activation/out/glu-situglu-100L_run/checkpoint-2100/rng_state.pth +3 -0
  32. zain/Activation/out/glu-situglu-100L_run/checkpoint-2100/scheduler.pt +3 -0
  33. zain/Activation/out/glu-situglu-100L_run/checkpoint-2100/tokenizer.json +0 -0
  34. zain/Activation/out/glu-situglu-100L_run/checkpoint-2100/tokenizer_config.json +13 -0
  35. zain/Activation/out/glu-situglu-100L_run/checkpoint-2100/trainer_state.json +769 -0
  36. zain/Activation/out/glu-situglu-100L_run/checkpoint-2100/training_args.bin +3 -0
  37. zain/Activation/out/glu-situglu-100L_run/checkpoint-2200/config.json +36 -0
  38. zain/Activation/out/glu-situglu-100L_run/checkpoint-2200/model.safetensors +3 -0
  39. zain/Activation/out/glu-situglu-100L_run/checkpoint-2200/optimizer.pt +3 -0
  40. zain/Activation/out/glu-situglu-100L_run/checkpoint-2200/rng_state.pth +3 -0
  41. zain/Activation/out/glu-situglu-100L_run/checkpoint-2200/scheduler.pt +3 -0
  42. zain/Activation/out/glu-situglu-100L_run/checkpoint-2200/tokenizer.json +0 -0
  43. zain/Activation/out/glu-situglu-100L_run/checkpoint-2200/tokenizer_config.json +13 -0
  44. zain/Activation/out/glu-situglu-100L_run/checkpoint-2200/trainer_state.json +804 -0
  45. zain/Activation/out/glu-situglu-100L_run/checkpoint-2200/training_args.bin +3 -0
  46. zain/Activation/out/glu-situglu-100L_run/checkpoint-2300/config.json +36 -0
  47. zain/Activation/out/glu-situglu-100L_run/checkpoint-2300/model.safetensors +3 -0
  48. zain/Activation/out/glu-situglu-100L_run/checkpoint-2300/optimizer.pt +3 -0
  49. zain/Activation/out/glu-situglu-100L_run/checkpoint-2300/rng_state.pth +3 -0
  50. zain/Activation/out/glu-situglu-100L_run/checkpoint-2300/scheduler.pt +3 -0
zain/Activation/out/glu-situglu-100L_run/checkpoint-1800/config.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation": "situglu",
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": 100,
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-situglu-100L_run/checkpoint-1800/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a69d017eb6f74bd9904bbda05bc3d1bf802701ce3dddc6b51a89580814b9e532
3
+ size 33967272
zain/Activation/out/glu-situglu-100L_run/checkpoint-1800/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a1b0e6ca0758580321b3677c983141dc920ffe3d31cca235b71e21c274aa3406
3
+ size 68504996
zain/Activation/out/glu-situglu-100L_run/checkpoint-1800/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-situglu-100L_run/checkpoint-1800/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8214bf4c471bc465884f1c219edb27abec42cbdf74a9a2fe9320000a4ce60c16
3
+ size 1064
zain/Activation/out/glu-situglu-100L_run/checkpoint-1800/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
zain/Activation/out/glu-situglu-100L_run/checkpoint-1800/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-situglu-100L_run/checkpoint-1800/trainer_state.json ADDED
@@ -0,0 +1,664 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 0.12133468149646107,
6
+ "eval_steps": 2498,
7
+ "global_step": 1800,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.0013481631277384564,
14
+ "grad_norm": 2.078125,
15
+ "learning_rate": 2.66e-05,
16
+ "loss": 8.299653625488281,
17
+ "step": 20
18
+ },
19
+ {
20
+ "epoch": 0.002696326255476913,
21
+ "grad_norm": 1.3515625,
22
+ "learning_rate": 5.46e-05,
23
+ "loss": 8.04769058227539,
24
+ "step": 40
25
+ },
26
+ {
27
+ "epoch": 0.004044489383215369,
28
+ "grad_norm": 1.2734375,
29
+ "learning_rate": 8.259999999999999e-05,
30
+ "loss": 7.8379150390625,
31
+ "step": 60
32
+ },
33
+ {
34
+ "epoch": 0.005392652510953826,
35
+ "grad_norm": 1.265625,
36
+ "learning_rate": 0.0001106,
37
+ "loss": 7.572837066650391,
38
+ "step": 80
39
+ },
40
+ {
41
+ "epoch": 0.006740815638692282,
42
+ "grad_norm": 1.1953125,
43
+ "learning_rate": 0.0001386,
44
+ "loss": 7.233182525634765,
45
+ "step": 100
46
+ },
47
+ {
48
+ "epoch": 0.008088978766430738,
49
+ "grad_norm": 1.1328125,
50
+ "learning_rate": 0.00016659999999999998,
51
+ "loss": 6.860689544677735,
52
+ "step": 120
53
+ },
54
+ {
55
+ "epoch": 0.009437141894169195,
56
+ "grad_norm": 0.94921875,
57
+ "learning_rate": 0.00019460000000000001,
58
+ "loss": 6.517751312255859,
59
+ "step": 140
60
+ },
61
+ {
62
+ "epoch": 0.010785305021907651,
63
+ "grad_norm": 0.8046875,
64
+ "learning_rate": 0.0002226,
65
+ "loss": 6.239968109130859,
66
+ "step": 160
67
+ },
68
+ {
69
+ "epoch": 0.012133468149646108,
70
+ "grad_norm": 0.6640625,
71
+ "learning_rate": 0.00025059999999999997,
72
+ "loss": 6.041232299804688,
73
+ "step": 180
74
+ },
75
+ {
76
+ "epoch": 0.013481631277384564,
77
+ "grad_norm": 0.5234375,
78
+ "learning_rate": 0.0002786,
79
+ "loss": 5.863973999023438,
80
+ "step": 200
81
+ },
82
+ {
83
+ "epoch": 0.01482979440512302,
84
+ "grad_norm": 1.375,
85
+ "learning_rate": 0.00030659999999999997,
86
+ "loss": 5.692941665649414,
87
+ "step": 220
88
+ },
89
+ {
90
+ "epoch": 0.016177957532861477,
91
+ "grad_norm": 0.875,
92
+ "learning_rate": 0.0003346,
93
+ "loss": 5.484311294555664,
94
+ "step": 240
95
+ },
96
+ {
97
+ "epoch": 0.01752612066059993,
98
+ "grad_norm": 1.25,
99
+ "learning_rate": 0.00036260000000000003,
100
+ "loss": 5.25721549987793,
101
+ "step": 260
102
+ },
103
+ {
104
+ "epoch": 0.01887428378833839,
105
+ "grad_norm": 0.8125,
106
+ "learning_rate": 0.0003906,
107
+ "loss": 5.028667831420899,
108
+ "step": 280
109
+ },
110
+ {
111
+ "epoch": 0.020222446916076844,
112
+ "grad_norm": 0.75390625,
113
+ "learning_rate": 0.0004186,
114
+ "loss": 4.791546249389649,
115
+ "step": 300
116
+ },
117
+ {
118
+ "epoch": 0.021570610043815303,
119
+ "grad_norm": 0.875,
120
+ "learning_rate": 0.0004466,
121
+ "loss": 4.598950958251953,
122
+ "step": 320
123
+ },
124
+ {
125
+ "epoch": 0.022918773171553757,
126
+ "grad_norm": 0.8125,
127
+ "learning_rate": 0.00047460000000000004,
128
+ "loss": 4.41942024230957,
129
+ "step": 340
130
+ },
131
+ {
132
+ "epoch": 0.024266936299292215,
133
+ "grad_norm": 0.890625,
134
+ "learning_rate": 0.0005026,
135
+ "loss": 4.27253303527832,
136
+ "step": 360
137
+ },
138
+ {
139
+ "epoch": 0.02561509942703067,
140
+ "grad_norm": 1.2734375,
141
+ "learning_rate": 0.0005306,
142
+ "loss": 4.1283222198486325,
143
+ "step": 380
144
+ },
145
+ {
146
+ "epoch": 0.026963262554769128,
147
+ "grad_norm": 0.54296875,
148
+ "learning_rate": 0.0005586,
149
+ "loss": 3.9935020446777343,
150
+ "step": 400
151
+ },
152
+ {
153
+ "epoch": 0.028311425682507583,
154
+ "grad_norm": 0.546875,
155
+ "learning_rate": 0.0005866,
156
+ "loss": 3.885255813598633,
157
+ "step": 420
158
+ },
159
+ {
160
+ "epoch": 0.02965958881024604,
161
+ "grad_norm": 0.7890625,
162
+ "learning_rate": 0.0006146,
163
+ "loss": 3.821950912475586,
164
+ "step": 440
165
+ },
166
+ {
167
+ "epoch": 0.031007751937984496,
168
+ "grad_norm": 0.52734375,
169
+ "learning_rate": 0.0006426,
170
+ "loss": 3.7068336486816404,
171
+ "step": 460
172
+ },
173
+ {
174
+ "epoch": 0.032355915065722954,
175
+ "grad_norm": 0.6328125,
176
+ "learning_rate": 0.0006705999999999999,
177
+ "loss": 3.6370723724365233,
178
+ "step": 480
179
+ },
180
+ {
181
+ "epoch": 0.03370407819346141,
182
+ "grad_norm": 0.5625,
183
+ "learning_rate": 0.0006986,
184
+ "loss": 3.5550304412841798,
185
+ "step": 500
186
+ },
187
+ {
188
+ "epoch": 0.03505224132119986,
189
+ "grad_norm": 0.47265625,
190
+ "learning_rate": 0.0007,
191
+ "loss": 3.4874305725097656,
192
+ "step": 520
193
+ },
194
+ {
195
+ "epoch": 0.03640040444893832,
196
+ "grad_norm": 0.6328125,
197
+ "learning_rate": 0.0007,
198
+ "loss": 3.414617919921875,
199
+ "step": 540
200
+ },
201
+ {
202
+ "epoch": 0.03774856757667678,
203
+ "grad_norm": 0.478515625,
204
+ "learning_rate": 0.0007,
205
+ "loss": 3.3621952056884767,
206
+ "step": 560
207
+ },
208
+ {
209
+ "epoch": 0.03909673070441524,
210
+ "grad_norm": 0.478515625,
211
+ "learning_rate": 0.0007,
212
+ "loss": 3.301148223876953,
213
+ "step": 580
214
+ },
215
+ {
216
+ "epoch": 0.04044489383215369,
217
+ "grad_norm": 0.44140625,
218
+ "learning_rate": 0.0007,
219
+ "loss": 3.266124725341797,
220
+ "step": 600
221
+ },
222
+ {
223
+ "epoch": 0.04179305695989215,
224
+ "grad_norm": 0.58203125,
225
+ "learning_rate": 0.0007,
226
+ "loss": 3.1991281509399414,
227
+ "step": 620
228
+ },
229
+ {
230
+ "epoch": 0.043141220087630605,
231
+ "grad_norm": 0.51953125,
232
+ "learning_rate": 0.0007,
233
+ "loss": 3.175183868408203,
234
+ "step": 640
235
+ },
236
+ {
237
+ "epoch": 0.044489383215369056,
238
+ "grad_norm": 0.443359375,
239
+ "learning_rate": 0.0007,
240
+ "loss": 3.1185443878173826,
241
+ "step": 660
242
+ },
243
+ {
244
+ "epoch": 0.045837546343107514,
245
+ "grad_norm": 0.447265625,
246
+ "learning_rate": 0.0007,
247
+ "loss": 3.091288757324219,
248
+ "step": 680
249
+ },
250
+ {
251
+ "epoch": 0.04718570947084597,
252
+ "grad_norm": 0.400390625,
253
+ "learning_rate": 0.0007,
254
+ "loss": 3.0479175567626955,
255
+ "step": 700
256
+ },
257
+ {
258
+ "epoch": 0.04853387259858443,
259
+ "grad_norm": 0.39453125,
260
+ "learning_rate": 0.0007,
261
+ "loss": 3.012004280090332,
262
+ "step": 720
263
+ },
264
+ {
265
+ "epoch": 0.04988203572632288,
266
+ "grad_norm": 0.40625,
267
+ "learning_rate": 0.0007,
268
+ "loss": 2.973306655883789,
269
+ "step": 740
270
+ },
271
+ {
272
+ "epoch": 0.05123019885406134,
273
+ "grad_norm": 0.41015625,
274
+ "learning_rate": 0.0007,
275
+ "loss": 2.9569366455078123,
276
+ "step": 760
277
+ },
278
+ {
279
+ "epoch": 0.0525783619817998,
280
+ "grad_norm": 0.427734375,
281
+ "learning_rate": 0.0007,
282
+ "loss": 2.959245491027832,
283
+ "step": 780
284
+ },
285
+ {
286
+ "epoch": 0.053926525109538256,
287
+ "grad_norm": 0.4296875,
288
+ "learning_rate": 0.0007,
289
+ "loss": 2.9073596954345704,
290
+ "step": 800
291
+ },
292
+ {
293
+ "epoch": 0.05527468823727671,
294
+ "grad_norm": 0.435546875,
295
+ "learning_rate": 0.0007,
296
+ "loss": 2.894051933288574,
297
+ "step": 820
298
+ },
299
+ {
300
+ "epoch": 0.056622851365015166,
301
+ "grad_norm": 0.4296875,
302
+ "learning_rate": 0.0007,
303
+ "loss": 2.86330623626709,
304
+ "step": 840
305
+ },
306
+ {
307
+ "epoch": 0.057971014492753624,
308
+ "grad_norm": 0.388671875,
309
+ "learning_rate": 0.0007,
310
+ "loss": 2.81943302154541,
311
+ "step": 860
312
+ },
313
+ {
314
+ "epoch": 0.05931917762049208,
315
+ "grad_norm": 0.384765625,
316
+ "learning_rate": 0.0007,
317
+ "loss": 2.7944929122924806,
318
+ "step": 880
319
+ },
320
+ {
321
+ "epoch": 0.06066734074823053,
322
+ "grad_norm": 0.40234375,
323
+ "learning_rate": 0.0007,
324
+ "loss": 2.779962730407715,
325
+ "step": 900
326
+ },
327
+ {
328
+ "epoch": 0.06201550387596899,
329
+ "grad_norm": 0.416015625,
330
+ "learning_rate": 0.0007,
331
+ "loss": 2.751751708984375,
332
+ "step": 920
333
+ },
334
+ {
335
+ "epoch": 0.06336366700370745,
336
+ "grad_norm": 0.447265625,
337
+ "learning_rate": 0.0007,
338
+ "loss": 2.7392576217651365,
339
+ "step": 940
340
+ },
341
+ {
342
+ "epoch": 0.06471183013144591,
343
+ "grad_norm": 0.44140625,
344
+ "learning_rate": 0.0007,
345
+ "loss": 2.7111114501953124,
346
+ "step": 960
347
+ },
348
+ {
349
+ "epoch": 0.06605999325918437,
350
+ "grad_norm": 0.39453125,
351
+ "learning_rate": 0.0007,
352
+ "loss": 2.692020797729492,
353
+ "step": 980
354
+ },
355
+ {
356
+ "epoch": 0.06740815638692282,
357
+ "grad_norm": 0.36328125,
358
+ "learning_rate": 0.0007,
359
+ "loss": 2.691050910949707,
360
+ "step": 1000
361
+ },
362
+ {
363
+ "epoch": 0.06875631951466127,
364
+ "grad_norm": 0.37109375,
365
+ "learning_rate": 0.0007,
366
+ "loss": 2.6589706420898436,
367
+ "step": 1020
368
+ },
369
+ {
370
+ "epoch": 0.07010448264239973,
371
+ "grad_norm": 0.40625,
372
+ "learning_rate": 0.0007,
373
+ "loss": 2.643506622314453,
374
+ "step": 1040
375
+ },
376
+ {
377
+ "epoch": 0.07145264577013818,
378
+ "grad_norm": 0.390625,
379
+ "learning_rate": 0.0007,
380
+ "loss": 2.629362869262695,
381
+ "step": 1060
382
+ },
383
+ {
384
+ "epoch": 0.07280080889787664,
385
+ "grad_norm": 0.37890625,
386
+ "learning_rate": 0.0007,
387
+ "loss": 2.606895637512207,
388
+ "step": 1080
389
+ },
390
+ {
391
+ "epoch": 0.0741489720256151,
392
+ "grad_norm": 0.412109375,
393
+ "learning_rate": 0.0007,
394
+ "loss": 2.6003129959106444,
395
+ "step": 1100
396
+ },
397
+ {
398
+ "epoch": 0.07549713515335356,
399
+ "grad_norm": 0.384765625,
400
+ "learning_rate": 0.0007,
401
+ "loss": 2.565888786315918,
402
+ "step": 1120
403
+ },
404
+ {
405
+ "epoch": 0.07684529828109202,
406
+ "grad_norm": 0.41796875,
407
+ "learning_rate": 0.0007,
408
+ "loss": 2.55434627532959,
409
+ "step": 1140
410
+ },
411
+ {
412
+ "epoch": 0.07819346140883048,
413
+ "grad_norm": 0.37890625,
414
+ "learning_rate": 0.0007,
415
+ "loss": 2.5499151229858397,
416
+ "step": 1160
417
+ },
418
+ {
419
+ "epoch": 0.07954162453656892,
420
+ "grad_norm": 0.38671875,
421
+ "learning_rate": 0.0007,
422
+ "loss": 2.5433088302612306,
423
+ "step": 1180
424
+ },
425
+ {
426
+ "epoch": 0.08088978766430738,
427
+ "grad_norm": 0.45703125,
428
+ "learning_rate": 0.0007,
429
+ "loss": 2.533991050720215,
430
+ "step": 1200
431
+ },
432
+ {
433
+ "epoch": 0.08223795079204584,
434
+ "grad_norm": 0.359375,
435
+ "learning_rate": 0.0007,
436
+ "loss": 2.5188695907592775,
437
+ "step": 1220
438
+ },
439
+ {
440
+ "epoch": 0.0835861139197843,
441
+ "grad_norm": 0.36328125,
442
+ "learning_rate": 0.0007,
443
+ "loss": 2.504729080200195,
444
+ "step": 1240
445
+ },
446
+ {
447
+ "epoch": 0.08493427704752275,
448
+ "grad_norm": 0.3671875,
449
+ "learning_rate": 0.0007,
450
+ "loss": 2.4896400451660154,
451
+ "step": 1260
452
+ },
453
+ {
454
+ "epoch": 0.08628244017526121,
455
+ "grad_norm": 0.369140625,
456
+ "learning_rate": 0.0007,
457
+ "loss": 2.4696292877197266,
458
+ "step": 1280
459
+ },
460
+ {
461
+ "epoch": 0.08763060330299967,
462
+ "grad_norm": 0.369140625,
463
+ "learning_rate": 0.0007,
464
+ "loss": 2.4618251800537108,
465
+ "step": 1300
466
+ },
467
+ {
468
+ "epoch": 0.08897876643073811,
469
+ "grad_norm": 0.361328125,
470
+ "learning_rate": 0.0007,
471
+ "loss": 2.454075813293457,
472
+ "step": 1320
473
+ },
474
+ {
475
+ "epoch": 0.09032692955847657,
476
+ "grad_norm": 0.365234375,
477
+ "learning_rate": 0.0007,
478
+ "loss": 2.441549301147461,
479
+ "step": 1340
480
+ },
481
+ {
482
+ "epoch": 0.09167509268621503,
483
+ "grad_norm": 0.408203125,
484
+ "learning_rate": 0.0007,
485
+ "loss": 2.440728950500488,
486
+ "step": 1360
487
+ },
488
+ {
489
+ "epoch": 0.09302325581395349,
490
+ "grad_norm": 0.369140625,
491
+ "learning_rate": 0.0007,
492
+ "loss": 2.417755889892578,
493
+ "step": 1380
494
+ },
495
+ {
496
+ "epoch": 0.09437141894169195,
497
+ "grad_norm": 0.34765625,
498
+ "learning_rate": 0.0007,
499
+ "loss": 2.408277893066406,
500
+ "step": 1400
501
+ },
502
+ {
503
+ "epoch": 0.0957195820694304,
504
+ "grad_norm": 0.380859375,
505
+ "learning_rate": 0.0007,
506
+ "loss": 2.4170982360839846,
507
+ "step": 1420
508
+ },
509
+ {
510
+ "epoch": 0.09706774519716886,
511
+ "grad_norm": 0.400390625,
512
+ "learning_rate": 0.0007,
513
+ "loss": 2.397001838684082,
514
+ "step": 1440
515
+ },
516
+ {
517
+ "epoch": 0.09841590832490732,
518
+ "grad_norm": 0.380859375,
519
+ "learning_rate": 0.0007,
520
+ "loss": 2.3847665786743164,
521
+ "step": 1460
522
+ },
523
+ {
524
+ "epoch": 0.09976407145264576,
525
+ "grad_norm": 0.357421875,
526
+ "learning_rate": 0.0007,
527
+ "loss": 2.3689859390258787,
528
+ "step": 1480
529
+ },
530
+ {
531
+ "epoch": 0.10111223458038422,
532
+ "grad_norm": 0.365234375,
533
+ "learning_rate": 0.0007,
534
+ "loss": 2.3572067260742187,
535
+ "step": 1500
536
+ },
537
+ {
538
+ "epoch": 0.10246039770812268,
539
+ "grad_norm": 0.359375,
540
+ "learning_rate": 0.0007,
541
+ "loss": 2.3587852478027345,
542
+ "step": 1520
543
+ },
544
+ {
545
+ "epoch": 0.10380856083586114,
546
+ "grad_norm": 0.3984375,
547
+ "learning_rate": 0.0007,
548
+ "loss": 2.3474098205566407,
549
+ "step": 1540
550
+ },
551
+ {
552
+ "epoch": 0.1051567239635996,
553
+ "grad_norm": 0.369140625,
554
+ "learning_rate": 0.0007,
555
+ "loss": 2.336527633666992,
556
+ "step": 1560
557
+ },
558
+ {
559
+ "epoch": 0.10650488709133805,
560
+ "grad_norm": 0.34765625,
561
+ "learning_rate": 0.0007,
562
+ "loss": 2.354816436767578,
563
+ "step": 1580
564
+ },
565
+ {
566
+ "epoch": 0.10785305021907651,
567
+ "grad_norm": 0.40625,
568
+ "learning_rate": 0.0007,
569
+ "loss": 2.3224559783935548,
570
+ "step": 1600
571
+ },
572
+ {
573
+ "epoch": 0.10920121334681497,
574
+ "grad_norm": 0.3515625,
575
+ "learning_rate": 0.0007,
576
+ "loss": 2.320570945739746,
577
+ "step": 1620
578
+ },
579
+ {
580
+ "epoch": 0.11054937647455342,
581
+ "grad_norm": 0.3515625,
582
+ "learning_rate": 0.0007,
583
+ "loss": 2.331516075134277,
584
+ "step": 1640
585
+ },
586
+ {
587
+ "epoch": 0.11189753960229187,
588
+ "grad_norm": 0.326171875,
589
+ "learning_rate": 0.0007,
590
+ "loss": 2.3128894805908202,
591
+ "step": 1660
592
+ },
593
+ {
594
+ "epoch": 0.11324570273003033,
595
+ "grad_norm": 0.341796875,
596
+ "learning_rate": 0.0007,
597
+ "loss": 2.2964305877685547,
598
+ "step": 1680
599
+ },
600
+ {
601
+ "epoch": 0.11459386585776879,
602
+ "grad_norm": 0.376953125,
603
+ "learning_rate": 0.0007,
604
+ "loss": 2.3081363677978515,
605
+ "step": 1700
606
+ },
607
+ {
608
+ "epoch": 0.11594202898550725,
609
+ "grad_norm": 0.375,
610
+ "learning_rate": 0.0007,
611
+ "loss": 2.2839227676391602,
612
+ "step": 1720
613
+ },
614
+ {
615
+ "epoch": 0.1172901921132457,
616
+ "grad_norm": 0.333984375,
617
+ "learning_rate": 0.0007,
618
+ "loss": 2.2729005813598633,
619
+ "step": 1740
620
+ },
621
+ {
622
+ "epoch": 0.11863835524098416,
623
+ "grad_norm": 0.34375,
624
+ "learning_rate": 0.0007,
625
+ "loss": 2.2774417877197264,
626
+ "step": 1760
627
+ },
628
+ {
629
+ "epoch": 0.11998651836872262,
630
+ "grad_norm": 0.390625,
631
+ "learning_rate": 0.0007,
632
+ "loss": 2.290260887145996,
633
+ "step": 1780
634
+ },
635
+ {
636
+ "epoch": 0.12133468149646107,
637
+ "grad_norm": 0.3515625,
638
+ "learning_rate": 0.0007,
639
+ "loss": 2.286345672607422,
640
+ "step": 1800
641
+ }
642
+ ],
643
+ "logging_steps": 20,
644
+ "max_steps": 2500,
645
+ "num_input_tokens_seen": 0,
646
+ "num_train_epochs": 1,
647
+ "save_steps": 100,
648
+ "stateful_callbacks": {
649
+ "TrainerControl": {
650
+ "args": {
651
+ "should_epoch_stop": false,
652
+ "should_evaluate": false,
653
+ "should_log": false,
654
+ "should_save": true,
655
+ "should_training_stop": false
656
+ },
657
+ "attributes": {}
658
+ }
659
+ },
660
+ "total_flos": 5807310844723200.0,
661
+ "train_batch_size": 64,
662
+ "trial_name": null,
663
+ "trial_params": null
664
+ }
zain/Activation/out/glu-situglu-100L_run/checkpoint-1800/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:26b99898fe48ff0e75b1f0139845f22400023db32e87c8fca3404fda1b772936
3
+ size 4920
zain/Activation/out/glu-situglu-100L_run/checkpoint-1900/config.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation": "situglu",
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": 100,
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-situglu-100L_run/checkpoint-1900/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:11cea425ef8d3f7751b7e1eac39b474946212c24b0c27274328178a71a422bf3
3
+ size 33967272
zain/Activation/out/glu-situglu-100L_run/checkpoint-1900/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9221d0f55b115450a25a68b3a0e7416cc13f6a41d70e545e8a55451cf36fbd12
3
+ size 68504996
zain/Activation/out/glu-situglu-100L_run/checkpoint-1900/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-situglu-100L_run/checkpoint-1900/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aa8588bc4a247e7962c4619ae7e7ba14fb542cc23b88b7c743a799982e027413
3
+ size 1064
zain/Activation/out/glu-situglu-100L_run/checkpoint-1900/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
zain/Activation/out/glu-situglu-100L_run/checkpoint-1900/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-situglu-100L_run/checkpoint-1900/trainer_state.json ADDED
@@ -0,0 +1,699 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 0.12807549713515334,
6
+ "eval_steps": 2498,
7
+ "global_step": 1900,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.0013481631277384564,
14
+ "grad_norm": 2.078125,
15
+ "learning_rate": 2.66e-05,
16
+ "loss": 8.299653625488281,
17
+ "step": 20
18
+ },
19
+ {
20
+ "epoch": 0.002696326255476913,
21
+ "grad_norm": 1.3515625,
22
+ "learning_rate": 5.46e-05,
23
+ "loss": 8.04769058227539,
24
+ "step": 40
25
+ },
26
+ {
27
+ "epoch": 0.004044489383215369,
28
+ "grad_norm": 1.2734375,
29
+ "learning_rate": 8.259999999999999e-05,
30
+ "loss": 7.8379150390625,
31
+ "step": 60
32
+ },
33
+ {
34
+ "epoch": 0.005392652510953826,
35
+ "grad_norm": 1.265625,
36
+ "learning_rate": 0.0001106,
37
+ "loss": 7.572837066650391,
38
+ "step": 80
39
+ },
40
+ {
41
+ "epoch": 0.006740815638692282,
42
+ "grad_norm": 1.1953125,
43
+ "learning_rate": 0.0001386,
44
+ "loss": 7.233182525634765,
45
+ "step": 100
46
+ },
47
+ {
48
+ "epoch": 0.008088978766430738,
49
+ "grad_norm": 1.1328125,
50
+ "learning_rate": 0.00016659999999999998,
51
+ "loss": 6.860689544677735,
52
+ "step": 120
53
+ },
54
+ {
55
+ "epoch": 0.009437141894169195,
56
+ "grad_norm": 0.94921875,
57
+ "learning_rate": 0.00019460000000000001,
58
+ "loss": 6.517751312255859,
59
+ "step": 140
60
+ },
61
+ {
62
+ "epoch": 0.010785305021907651,
63
+ "grad_norm": 0.8046875,
64
+ "learning_rate": 0.0002226,
65
+ "loss": 6.239968109130859,
66
+ "step": 160
67
+ },
68
+ {
69
+ "epoch": 0.012133468149646108,
70
+ "grad_norm": 0.6640625,
71
+ "learning_rate": 0.00025059999999999997,
72
+ "loss": 6.041232299804688,
73
+ "step": 180
74
+ },
75
+ {
76
+ "epoch": 0.013481631277384564,
77
+ "grad_norm": 0.5234375,
78
+ "learning_rate": 0.0002786,
79
+ "loss": 5.863973999023438,
80
+ "step": 200
81
+ },
82
+ {
83
+ "epoch": 0.01482979440512302,
84
+ "grad_norm": 1.375,
85
+ "learning_rate": 0.00030659999999999997,
86
+ "loss": 5.692941665649414,
87
+ "step": 220
88
+ },
89
+ {
90
+ "epoch": 0.016177957532861477,
91
+ "grad_norm": 0.875,
92
+ "learning_rate": 0.0003346,
93
+ "loss": 5.484311294555664,
94
+ "step": 240
95
+ },
96
+ {
97
+ "epoch": 0.01752612066059993,
98
+ "grad_norm": 1.25,
99
+ "learning_rate": 0.00036260000000000003,
100
+ "loss": 5.25721549987793,
101
+ "step": 260
102
+ },
103
+ {
104
+ "epoch": 0.01887428378833839,
105
+ "grad_norm": 0.8125,
106
+ "learning_rate": 0.0003906,
107
+ "loss": 5.028667831420899,
108
+ "step": 280
109
+ },
110
+ {
111
+ "epoch": 0.020222446916076844,
112
+ "grad_norm": 0.75390625,
113
+ "learning_rate": 0.0004186,
114
+ "loss": 4.791546249389649,
115
+ "step": 300
116
+ },
117
+ {
118
+ "epoch": 0.021570610043815303,
119
+ "grad_norm": 0.875,
120
+ "learning_rate": 0.0004466,
121
+ "loss": 4.598950958251953,
122
+ "step": 320
123
+ },
124
+ {
125
+ "epoch": 0.022918773171553757,
126
+ "grad_norm": 0.8125,
127
+ "learning_rate": 0.00047460000000000004,
128
+ "loss": 4.41942024230957,
129
+ "step": 340
130
+ },
131
+ {
132
+ "epoch": 0.024266936299292215,
133
+ "grad_norm": 0.890625,
134
+ "learning_rate": 0.0005026,
135
+ "loss": 4.27253303527832,
136
+ "step": 360
137
+ },
138
+ {
139
+ "epoch": 0.02561509942703067,
140
+ "grad_norm": 1.2734375,
141
+ "learning_rate": 0.0005306,
142
+ "loss": 4.1283222198486325,
143
+ "step": 380
144
+ },
145
+ {
146
+ "epoch": 0.026963262554769128,
147
+ "grad_norm": 0.54296875,
148
+ "learning_rate": 0.0005586,
149
+ "loss": 3.9935020446777343,
150
+ "step": 400
151
+ },
152
+ {
153
+ "epoch": 0.028311425682507583,
154
+ "grad_norm": 0.546875,
155
+ "learning_rate": 0.0005866,
156
+ "loss": 3.885255813598633,
157
+ "step": 420
158
+ },
159
+ {
160
+ "epoch": 0.02965958881024604,
161
+ "grad_norm": 0.7890625,
162
+ "learning_rate": 0.0006146,
163
+ "loss": 3.821950912475586,
164
+ "step": 440
165
+ },
166
+ {
167
+ "epoch": 0.031007751937984496,
168
+ "grad_norm": 0.52734375,
169
+ "learning_rate": 0.0006426,
170
+ "loss": 3.7068336486816404,
171
+ "step": 460
172
+ },
173
+ {
174
+ "epoch": 0.032355915065722954,
175
+ "grad_norm": 0.6328125,
176
+ "learning_rate": 0.0006705999999999999,
177
+ "loss": 3.6370723724365233,
178
+ "step": 480
179
+ },
180
+ {
181
+ "epoch": 0.03370407819346141,
182
+ "grad_norm": 0.5625,
183
+ "learning_rate": 0.0006986,
184
+ "loss": 3.5550304412841798,
185
+ "step": 500
186
+ },
187
+ {
188
+ "epoch": 0.03505224132119986,
189
+ "grad_norm": 0.47265625,
190
+ "learning_rate": 0.0007,
191
+ "loss": 3.4874305725097656,
192
+ "step": 520
193
+ },
194
+ {
195
+ "epoch": 0.03640040444893832,
196
+ "grad_norm": 0.6328125,
197
+ "learning_rate": 0.0007,
198
+ "loss": 3.414617919921875,
199
+ "step": 540
200
+ },
201
+ {
202
+ "epoch": 0.03774856757667678,
203
+ "grad_norm": 0.478515625,
204
+ "learning_rate": 0.0007,
205
+ "loss": 3.3621952056884767,
206
+ "step": 560
207
+ },
208
+ {
209
+ "epoch": 0.03909673070441524,
210
+ "grad_norm": 0.478515625,
211
+ "learning_rate": 0.0007,
212
+ "loss": 3.301148223876953,
213
+ "step": 580
214
+ },
215
+ {
216
+ "epoch": 0.04044489383215369,
217
+ "grad_norm": 0.44140625,
218
+ "learning_rate": 0.0007,
219
+ "loss": 3.266124725341797,
220
+ "step": 600
221
+ },
222
+ {
223
+ "epoch": 0.04179305695989215,
224
+ "grad_norm": 0.58203125,
225
+ "learning_rate": 0.0007,
226
+ "loss": 3.1991281509399414,
227
+ "step": 620
228
+ },
229
+ {
230
+ "epoch": 0.043141220087630605,
231
+ "grad_norm": 0.51953125,
232
+ "learning_rate": 0.0007,
233
+ "loss": 3.175183868408203,
234
+ "step": 640
235
+ },
236
+ {
237
+ "epoch": 0.044489383215369056,
238
+ "grad_norm": 0.443359375,
239
+ "learning_rate": 0.0007,
240
+ "loss": 3.1185443878173826,
241
+ "step": 660
242
+ },
243
+ {
244
+ "epoch": 0.045837546343107514,
245
+ "grad_norm": 0.447265625,
246
+ "learning_rate": 0.0007,
247
+ "loss": 3.091288757324219,
248
+ "step": 680
249
+ },
250
+ {
251
+ "epoch": 0.04718570947084597,
252
+ "grad_norm": 0.400390625,
253
+ "learning_rate": 0.0007,
254
+ "loss": 3.0479175567626955,
255
+ "step": 700
256
+ },
257
+ {
258
+ "epoch": 0.04853387259858443,
259
+ "grad_norm": 0.39453125,
260
+ "learning_rate": 0.0007,
261
+ "loss": 3.012004280090332,
262
+ "step": 720
263
+ },
264
+ {
265
+ "epoch": 0.04988203572632288,
266
+ "grad_norm": 0.40625,
267
+ "learning_rate": 0.0007,
268
+ "loss": 2.973306655883789,
269
+ "step": 740
270
+ },
271
+ {
272
+ "epoch": 0.05123019885406134,
273
+ "grad_norm": 0.41015625,
274
+ "learning_rate": 0.0007,
275
+ "loss": 2.9569366455078123,
276
+ "step": 760
277
+ },
278
+ {
279
+ "epoch": 0.0525783619817998,
280
+ "grad_norm": 0.427734375,
281
+ "learning_rate": 0.0007,
282
+ "loss": 2.959245491027832,
283
+ "step": 780
284
+ },
285
+ {
286
+ "epoch": 0.053926525109538256,
287
+ "grad_norm": 0.4296875,
288
+ "learning_rate": 0.0007,
289
+ "loss": 2.9073596954345704,
290
+ "step": 800
291
+ },
292
+ {
293
+ "epoch": 0.05527468823727671,
294
+ "grad_norm": 0.435546875,
295
+ "learning_rate": 0.0007,
296
+ "loss": 2.894051933288574,
297
+ "step": 820
298
+ },
299
+ {
300
+ "epoch": 0.056622851365015166,
301
+ "grad_norm": 0.4296875,
302
+ "learning_rate": 0.0007,
303
+ "loss": 2.86330623626709,
304
+ "step": 840
305
+ },
306
+ {
307
+ "epoch": 0.057971014492753624,
308
+ "grad_norm": 0.388671875,
309
+ "learning_rate": 0.0007,
310
+ "loss": 2.81943302154541,
311
+ "step": 860
312
+ },
313
+ {
314
+ "epoch": 0.05931917762049208,
315
+ "grad_norm": 0.384765625,
316
+ "learning_rate": 0.0007,
317
+ "loss": 2.7944929122924806,
318
+ "step": 880
319
+ },
320
+ {
321
+ "epoch": 0.06066734074823053,
322
+ "grad_norm": 0.40234375,
323
+ "learning_rate": 0.0007,
324
+ "loss": 2.779962730407715,
325
+ "step": 900
326
+ },
327
+ {
328
+ "epoch": 0.06201550387596899,
329
+ "grad_norm": 0.416015625,
330
+ "learning_rate": 0.0007,
331
+ "loss": 2.751751708984375,
332
+ "step": 920
333
+ },
334
+ {
335
+ "epoch": 0.06336366700370745,
336
+ "grad_norm": 0.447265625,
337
+ "learning_rate": 0.0007,
338
+ "loss": 2.7392576217651365,
339
+ "step": 940
340
+ },
341
+ {
342
+ "epoch": 0.06471183013144591,
343
+ "grad_norm": 0.44140625,
344
+ "learning_rate": 0.0007,
345
+ "loss": 2.7111114501953124,
346
+ "step": 960
347
+ },
348
+ {
349
+ "epoch": 0.06605999325918437,
350
+ "grad_norm": 0.39453125,
351
+ "learning_rate": 0.0007,
352
+ "loss": 2.692020797729492,
353
+ "step": 980
354
+ },
355
+ {
356
+ "epoch": 0.06740815638692282,
357
+ "grad_norm": 0.36328125,
358
+ "learning_rate": 0.0007,
359
+ "loss": 2.691050910949707,
360
+ "step": 1000
361
+ },
362
+ {
363
+ "epoch": 0.06875631951466127,
364
+ "grad_norm": 0.37109375,
365
+ "learning_rate": 0.0007,
366
+ "loss": 2.6589706420898436,
367
+ "step": 1020
368
+ },
369
+ {
370
+ "epoch": 0.07010448264239973,
371
+ "grad_norm": 0.40625,
372
+ "learning_rate": 0.0007,
373
+ "loss": 2.643506622314453,
374
+ "step": 1040
375
+ },
376
+ {
377
+ "epoch": 0.07145264577013818,
378
+ "grad_norm": 0.390625,
379
+ "learning_rate": 0.0007,
380
+ "loss": 2.629362869262695,
381
+ "step": 1060
382
+ },
383
+ {
384
+ "epoch": 0.07280080889787664,
385
+ "grad_norm": 0.37890625,
386
+ "learning_rate": 0.0007,
387
+ "loss": 2.606895637512207,
388
+ "step": 1080
389
+ },
390
+ {
391
+ "epoch": 0.0741489720256151,
392
+ "grad_norm": 0.412109375,
393
+ "learning_rate": 0.0007,
394
+ "loss": 2.6003129959106444,
395
+ "step": 1100
396
+ },
397
+ {
398
+ "epoch": 0.07549713515335356,
399
+ "grad_norm": 0.384765625,
400
+ "learning_rate": 0.0007,
401
+ "loss": 2.565888786315918,
402
+ "step": 1120
403
+ },
404
+ {
405
+ "epoch": 0.07684529828109202,
406
+ "grad_norm": 0.41796875,
407
+ "learning_rate": 0.0007,
408
+ "loss": 2.55434627532959,
409
+ "step": 1140
410
+ },
411
+ {
412
+ "epoch": 0.07819346140883048,
413
+ "grad_norm": 0.37890625,
414
+ "learning_rate": 0.0007,
415
+ "loss": 2.5499151229858397,
416
+ "step": 1160
417
+ },
418
+ {
419
+ "epoch": 0.07954162453656892,
420
+ "grad_norm": 0.38671875,
421
+ "learning_rate": 0.0007,
422
+ "loss": 2.5433088302612306,
423
+ "step": 1180
424
+ },
425
+ {
426
+ "epoch": 0.08088978766430738,
427
+ "grad_norm": 0.45703125,
428
+ "learning_rate": 0.0007,
429
+ "loss": 2.533991050720215,
430
+ "step": 1200
431
+ },
432
+ {
433
+ "epoch": 0.08223795079204584,
434
+ "grad_norm": 0.359375,
435
+ "learning_rate": 0.0007,
436
+ "loss": 2.5188695907592775,
437
+ "step": 1220
438
+ },
439
+ {
440
+ "epoch": 0.0835861139197843,
441
+ "grad_norm": 0.36328125,
442
+ "learning_rate": 0.0007,
443
+ "loss": 2.504729080200195,
444
+ "step": 1240
445
+ },
446
+ {
447
+ "epoch": 0.08493427704752275,
448
+ "grad_norm": 0.3671875,
449
+ "learning_rate": 0.0007,
450
+ "loss": 2.4896400451660154,
451
+ "step": 1260
452
+ },
453
+ {
454
+ "epoch": 0.08628244017526121,
455
+ "grad_norm": 0.369140625,
456
+ "learning_rate": 0.0007,
457
+ "loss": 2.4696292877197266,
458
+ "step": 1280
459
+ },
460
+ {
461
+ "epoch": 0.08763060330299967,
462
+ "grad_norm": 0.369140625,
463
+ "learning_rate": 0.0007,
464
+ "loss": 2.4618251800537108,
465
+ "step": 1300
466
+ },
467
+ {
468
+ "epoch": 0.08897876643073811,
469
+ "grad_norm": 0.361328125,
470
+ "learning_rate": 0.0007,
471
+ "loss": 2.454075813293457,
472
+ "step": 1320
473
+ },
474
+ {
475
+ "epoch": 0.09032692955847657,
476
+ "grad_norm": 0.365234375,
477
+ "learning_rate": 0.0007,
478
+ "loss": 2.441549301147461,
479
+ "step": 1340
480
+ },
481
+ {
482
+ "epoch": 0.09167509268621503,
483
+ "grad_norm": 0.408203125,
484
+ "learning_rate": 0.0007,
485
+ "loss": 2.440728950500488,
486
+ "step": 1360
487
+ },
488
+ {
489
+ "epoch": 0.09302325581395349,
490
+ "grad_norm": 0.369140625,
491
+ "learning_rate": 0.0007,
492
+ "loss": 2.417755889892578,
493
+ "step": 1380
494
+ },
495
+ {
496
+ "epoch": 0.09437141894169195,
497
+ "grad_norm": 0.34765625,
498
+ "learning_rate": 0.0007,
499
+ "loss": 2.408277893066406,
500
+ "step": 1400
501
+ },
502
+ {
503
+ "epoch": 0.0957195820694304,
504
+ "grad_norm": 0.380859375,
505
+ "learning_rate": 0.0007,
506
+ "loss": 2.4170982360839846,
507
+ "step": 1420
508
+ },
509
+ {
510
+ "epoch": 0.09706774519716886,
511
+ "grad_norm": 0.400390625,
512
+ "learning_rate": 0.0007,
513
+ "loss": 2.397001838684082,
514
+ "step": 1440
515
+ },
516
+ {
517
+ "epoch": 0.09841590832490732,
518
+ "grad_norm": 0.380859375,
519
+ "learning_rate": 0.0007,
520
+ "loss": 2.3847665786743164,
521
+ "step": 1460
522
+ },
523
+ {
524
+ "epoch": 0.09976407145264576,
525
+ "grad_norm": 0.357421875,
526
+ "learning_rate": 0.0007,
527
+ "loss": 2.3689859390258787,
528
+ "step": 1480
529
+ },
530
+ {
531
+ "epoch": 0.10111223458038422,
532
+ "grad_norm": 0.365234375,
533
+ "learning_rate": 0.0007,
534
+ "loss": 2.3572067260742187,
535
+ "step": 1500
536
+ },
537
+ {
538
+ "epoch": 0.10246039770812268,
539
+ "grad_norm": 0.359375,
540
+ "learning_rate": 0.0007,
541
+ "loss": 2.3587852478027345,
542
+ "step": 1520
543
+ },
544
+ {
545
+ "epoch": 0.10380856083586114,
546
+ "grad_norm": 0.3984375,
547
+ "learning_rate": 0.0007,
548
+ "loss": 2.3474098205566407,
549
+ "step": 1540
550
+ },
551
+ {
552
+ "epoch": 0.1051567239635996,
553
+ "grad_norm": 0.369140625,
554
+ "learning_rate": 0.0007,
555
+ "loss": 2.336527633666992,
556
+ "step": 1560
557
+ },
558
+ {
559
+ "epoch": 0.10650488709133805,
560
+ "grad_norm": 0.34765625,
561
+ "learning_rate": 0.0007,
562
+ "loss": 2.354816436767578,
563
+ "step": 1580
564
+ },
565
+ {
566
+ "epoch": 0.10785305021907651,
567
+ "grad_norm": 0.40625,
568
+ "learning_rate": 0.0007,
569
+ "loss": 2.3224559783935548,
570
+ "step": 1600
571
+ },
572
+ {
573
+ "epoch": 0.10920121334681497,
574
+ "grad_norm": 0.3515625,
575
+ "learning_rate": 0.0007,
576
+ "loss": 2.320570945739746,
577
+ "step": 1620
578
+ },
579
+ {
580
+ "epoch": 0.11054937647455342,
581
+ "grad_norm": 0.3515625,
582
+ "learning_rate": 0.0007,
583
+ "loss": 2.331516075134277,
584
+ "step": 1640
585
+ },
586
+ {
587
+ "epoch": 0.11189753960229187,
588
+ "grad_norm": 0.326171875,
589
+ "learning_rate": 0.0007,
590
+ "loss": 2.3128894805908202,
591
+ "step": 1660
592
+ },
593
+ {
594
+ "epoch": 0.11324570273003033,
595
+ "grad_norm": 0.341796875,
596
+ "learning_rate": 0.0007,
597
+ "loss": 2.2964305877685547,
598
+ "step": 1680
599
+ },
600
+ {
601
+ "epoch": 0.11459386585776879,
602
+ "grad_norm": 0.376953125,
603
+ "learning_rate": 0.0007,
604
+ "loss": 2.3081363677978515,
605
+ "step": 1700
606
+ },
607
+ {
608
+ "epoch": 0.11594202898550725,
609
+ "grad_norm": 0.375,
610
+ "learning_rate": 0.0007,
611
+ "loss": 2.2839227676391602,
612
+ "step": 1720
613
+ },
614
+ {
615
+ "epoch": 0.1172901921132457,
616
+ "grad_norm": 0.333984375,
617
+ "learning_rate": 0.0007,
618
+ "loss": 2.2729005813598633,
619
+ "step": 1740
620
+ },
621
+ {
622
+ "epoch": 0.11863835524098416,
623
+ "grad_norm": 0.34375,
624
+ "learning_rate": 0.0007,
625
+ "loss": 2.2774417877197264,
626
+ "step": 1760
627
+ },
628
+ {
629
+ "epoch": 0.11998651836872262,
630
+ "grad_norm": 0.390625,
631
+ "learning_rate": 0.0007,
632
+ "loss": 2.290260887145996,
633
+ "step": 1780
634
+ },
635
+ {
636
+ "epoch": 0.12133468149646107,
637
+ "grad_norm": 0.3515625,
638
+ "learning_rate": 0.0007,
639
+ "loss": 2.286345672607422,
640
+ "step": 1800
641
+ },
642
+ {
643
+ "epoch": 0.12268284462419952,
644
+ "grad_norm": 0.396484375,
645
+ "learning_rate": 0.0007,
646
+ "loss": 2.278194236755371,
647
+ "step": 1820
648
+ },
649
+ {
650
+ "epoch": 0.12403100775193798,
651
+ "grad_norm": 0.345703125,
652
+ "learning_rate": 0.0007,
653
+ "loss": 2.245047760009766,
654
+ "step": 1840
655
+ },
656
+ {
657
+ "epoch": 0.12537917087967643,
658
+ "grad_norm": 0.369140625,
659
+ "learning_rate": 0.0007,
660
+ "loss": 2.2449272155761717,
661
+ "step": 1860
662
+ },
663
+ {
664
+ "epoch": 0.1267273340074149,
665
+ "grad_norm": 0.32421875,
666
+ "learning_rate": 0.0007,
667
+ "loss": 2.236011505126953,
668
+ "step": 1880
669
+ },
670
+ {
671
+ "epoch": 0.12807549713515334,
672
+ "grad_norm": 0.353515625,
673
+ "learning_rate": 0.0007,
674
+ "loss": 2.2427194595336912,
675
+ "step": 1900
676
+ }
677
+ ],
678
+ "logging_steps": 20,
679
+ "max_steps": 2500,
680
+ "num_input_tokens_seen": 0,
681
+ "num_train_epochs": 1,
682
+ "save_steps": 100,
683
+ "stateful_callbacks": {
684
+ "TrainerControl": {
685
+ "args": {
686
+ "should_epoch_stop": false,
687
+ "should_evaluate": false,
688
+ "should_log": false,
689
+ "should_save": true,
690
+ "should_training_stop": false
691
+ },
692
+ "attributes": {}
693
+ }
694
+ },
695
+ "total_flos": 6129939224985600.0,
696
+ "train_batch_size": 64,
697
+ "trial_name": null,
698
+ "trial_params": null
699
+ }
zain/Activation/out/glu-situglu-100L_run/checkpoint-1900/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:26b99898fe48ff0e75b1f0139845f22400023db32e87c8fca3404fda1b772936
3
+ size 4920
zain/Activation/out/glu-situglu-100L_run/checkpoint-2000/config.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation": "situglu",
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": 100,
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-situglu-100L_run/checkpoint-2000/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1c31833717b716d6b26ab8a3d5a74169cc6f5cc135c9c3922f255ef124380859
3
+ size 33967272
zain/Activation/out/glu-situglu-100L_run/checkpoint-2000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6ddc28eac85b431f50d10af631b372aa2373d536331f85e35f6ca5deb737bc8f
3
+ size 68504996
zain/Activation/out/glu-situglu-100L_run/checkpoint-2000/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-situglu-100L_run/checkpoint-2000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eb897171d6314e0d5ff47840624a03ce1078f22b8f8c3062289ef88ef22a406e
3
+ size 1064
zain/Activation/out/glu-situglu-100L_run/checkpoint-2000/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
zain/Activation/out/glu-situglu-100L_run/checkpoint-2000/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-situglu-100L_run/checkpoint-2000/trainer_state.json ADDED
@@ -0,0 +1,734 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 0.13481631277384565,
6
+ "eval_steps": 2498,
7
+ "global_step": 2000,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.0013481631277384564,
14
+ "grad_norm": 2.078125,
15
+ "learning_rate": 2.66e-05,
16
+ "loss": 8.299653625488281,
17
+ "step": 20
18
+ },
19
+ {
20
+ "epoch": 0.002696326255476913,
21
+ "grad_norm": 1.3515625,
22
+ "learning_rate": 5.46e-05,
23
+ "loss": 8.04769058227539,
24
+ "step": 40
25
+ },
26
+ {
27
+ "epoch": 0.004044489383215369,
28
+ "grad_norm": 1.2734375,
29
+ "learning_rate": 8.259999999999999e-05,
30
+ "loss": 7.8379150390625,
31
+ "step": 60
32
+ },
33
+ {
34
+ "epoch": 0.005392652510953826,
35
+ "grad_norm": 1.265625,
36
+ "learning_rate": 0.0001106,
37
+ "loss": 7.572837066650391,
38
+ "step": 80
39
+ },
40
+ {
41
+ "epoch": 0.006740815638692282,
42
+ "grad_norm": 1.1953125,
43
+ "learning_rate": 0.0001386,
44
+ "loss": 7.233182525634765,
45
+ "step": 100
46
+ },
47
+ {
48
+ "epoch": 0.008088978766430738,
49
+ "grad_norm": 1.1328125,
50
+ "learning_rate": 0.00016659999999999998,
51
+ "loss": 6.860689544677735,
52
+ "step": 120
53
+ },
54
+ {
55
+ "epoch": 0.009437141894169195,
56
+ "grad_norm": 0.94921875,
57
+ "learning_rate": 0.00019460000000000001,
58
+ "loss": 6.517751312255859,
59
+ "step": 140
60
+ },
61
+ {
62
+ "epoch": 0.010785305021907651,
63
+ "grad_norm": 0.8046875,
64
+ "learning_rate": 0.0002226,
65
+ "loss": 6.239968109130859,
66
+ "step": 160
67
+ },
68
+ {
69
+ "epoch": 0.012133468149646108,
70
+ "grad_norm": 0.6640625,
71
+ "learning_rate": 0.00025059999999999997,
72
+ "loss": 6.041232299804688,
73
+ "step": 180
74
+ },
75
+ {
76
+ "epoch": 0.013481631277384564,
77
+ "grad_norm": 0.5234375,
78
+ "learning_rate": 0.0002786,
79
+ "loss": 5.863973999023438,
80
+ "step": 200
81
+ },
82
+ {
83
+ "epoch": 0.01482979440512302,
84
+ "grad_norm": 1.375,
85
+ "learning_rate": 0.00030659999999999997,
86
+ "loss": 5.692941665649414,
87
+ "step": 220
88
+ },
89
+ {
90
+ "epoch": 0.016177957532861477,
91
+ "grad_norm": 0.875,
92
+ "learning_rate": 0.0003346,
93
+ "loss": 5.484311294555664,
94
+ "step": 240
95
+ },
96
+ {
97
+ "epoch": 0.01752612066059993,
98
+ "grad_norm": 1.25,
99
+ "learning_rate": 0.00036260000000000003,
100
+ "loss": 5.25721549987793,
101
+ "step": 260
102
+ },
103
+ {
104
+ "epoch": 0.01887428378833839,
105
+ "grad_norm": 0.8125,
106
+ "learning_rate": 0.0003906,
107
+ "loss": 5.028667831420899,
108
+ "step": 280
109
+ },
110
+ {
111
+ "epoch": 0.020222446916076844,
112
+ "grad_norm": 0.75390625,
113
+ "learning_rate": 0.0004186,
114
+ "loss": 4.791546249389649,
115
+ "step": 300
116
+ },
117
+ {
118
+ "epoch": 0.021570610043815303,
119
+ "grad_norm": 0.875,
120
+ "learning_rate": 0.0004466,
121
+ "loss": 4.598950958251953,
122
+ "step": 320
123
+ },
124
+ {
125
+ "epoch": 0.022918773171553757,
126
+ "grad_norm": 0.8125,
127
+ "learning_rate": 0.00047460000000000004,
128
+ "loss": 4.41942024230957,
129
+ "step": 340
130
+ },
131
+ {
132
+ "epoch": 0.024266936299292215,
133
+ "grad_norm": 0.890625,
134
+ "learning_rate": 0.0005026,
135
+ "loss": 4.27253303527832,
136
+ "step": 360
137
+ },
138
+ {
139
+ "epoch": 0.02561509942703067,
140
+ "grad_norm": 1.2734375,
141
+ "learning_rate": 0.0005306,
142
+ "loss": 4.1283222198486325,
143
+ "step": 380
144
+ },
145
+ {
146
+ "epoch": 0.026963262554769128,
147
+ "grad_norm": 0.54296875,
148
+ "learning_rate": 0.0005586,
149
+ "loss": 3.9935020446777343,
150
+ "step": 400
151
+ },
152
+ {
153
+ "epoch": 0.028311425682507583,
154
+ "grad_norm": 0.546875,
155
+ "learning_rate": 0.0005866,
156
+ "loss": 3.885255813598633,
157
+ "step": 420
158
+ },
159
+ {
160
+ "epoch": 0.02965958881024604,
161
+ "grad_norm": 0.7890625,
162
+ "learning_rate": 0.0006146,
163
+ "loss": 3.821950912475586,
164
+ "step": 440
165
+ },
166
+ {
167
+ "epoch": 0.031007751937984496,
168
+ "grad_norm": 0.52734375,
169
+ "learning_rate": 0.0006426,
170
+ "loss": 3.7068336486816404,
171
+ "step": 460
172
+ },
173
+ {
174
+ "epoch": 0.032355915065722954,
175
+ "grad_norm": 0.6328125,
176
+ "learning_rate": 0.0006705999999999999,
177
+ "loss": 3.6370723724365233,
178
+ "step": 480
179
+ },
180
+ {
181
+ "epoch": 0.03370407819346141,
182
+ "grad_norm": 0.5625,
183
+ "learning_rate": 0.0006986,
184
+ "loss": 3.5550304412841798,
185
+ "step": 500
186
+ },
187
+ {
188
+ "epoch": 0.03505224132119986,
189
+ "grad_norm": 0.47265625,
190
+ "learning_rate": 0.0007,
191
+ "loss": 3.4874305725097656,
192
+ "step": 520
193
+ },
194
+ {
195
+ "epoch": 0.03640040444893832,
196
+ "grad_norm": 0.6328125,
197
+ "learning_rate": 0.0007,
198
+ "loss": 3.414617919921875,
199
+ "step": 540
200
+ },
201
+ {
202
+ "epoch": 0.03774856757667678,
203
+ "grad_norm": 0.478515625,
204
+ "learning_rate": 0.0007,
205
+ "loss": 3.3621952056884767,
206
+ "step": 560
207
+ },
208
+ {
209
+ "epoch": 0.03909673070441524,
210
+ "grad_norm": 0.478515625,
211
+ "learning_rate": 0.0007,
212
+ "loss": 3.301148223876953,
213
+ "step": 580
214
+ },
215
+ {
216
+ "epoch": 0.04044489383215369,
217
+ "grad_norm": 0.44140625,
218
+ "learning_rate": 0.0007,
219
+ "loss": 3.266124725341797,
220
+ "step": 600
221
+ },
222
+ {
223
+ "epoch": 0.04179305695989215,
224
+ "grad_norm": 0.58203125,
225
+ "learning_rate": 0.0007,
226
+ "loss": 3.1991281509399414,
227
+ "step": 620
228
+ },
229
+ {
230
+ "epoch": 0.043141220087630605,
231
+ "grad_norm": 0.51953125,
232
+ "learning_rate": 0.0007,
233
+ "loss": 3.175183868408203,
234
+ "step": 640
235
+ },
236
+ {
237
+ "epoch": 0.044489383215369056,
238
+ "grad_norm": 0.443359375,
239
+ "learning_rate": 0.0007,
240
+ "loss": 3.1185443878173826,
241
+ "step": 660
242
+ },
243
+ {
244
+ "epoch": 0.045837546343107514,
245
+ "grad_norm": 0.447265625,
246
+ "learning_rate": 0.0007,
247
+ "loss": 3.091288757324219,
248
+ "step": 680
249
+ },
250
+ {
251
+ "epoch": 0.04718570947084597,
252
+ "grad_norm": 0.400390625,
253
+ "learning_rate": 0.0007,
254
+ "loss": 3.0479175567626955,
255
+ "step": 700
256
+ },
257
+ {
258
+ "epoch": 0.04853387259858443,
259
+ "grad_norm": 0.39453125,
260
+ "learning_rate": 0.0007,
261
+ "loss": 3.012004280090332,
262
+ "step": 720
263
+ },
264
+ {
265
+ "epoch": 0.04988203572632288,
266
+ "grad_norm": 0.40625,
267
+ "learning_rate": 0.0007,
268
+ "loss": 2.973306655883789,
269
+ "step": 740
270
+ },
271
+ {
272
+ "epoch": 0.05123019885406134,
273
+ "grad_norm": 0.41015625,
274
+ "learning_rate": 0.0007,
275
+ "loss": 2.9569366455078123,
276
+ "step": 760
277
+ },
278
+ {
279
+ "epoch": 0.0525783619817998,
280
+ "grad_norm": 0.427734375,
281
+ "learning_rate": 0.0007,
282
+ "loss": 2.959245491027832,
283
+ "step": 780
284
+ },
285
+ {
286
+ "epoch": 0.053926525109538256,
287
+ "grad_norm": 0.4296875,
288
+ "learning_rate": 0.0007,
289
+ "loss": 2.9073596954345704,
290
+ "step": 800
291
+ },
292
+ {
293
+ "epoch": 0.05527468823727671,
294
+ "grad_norm": 0.435546875,
295
+ "learning_rate": 0.0007,
296
+ "loss": 2.894051933288574,
297
+ "step": 820
298
+ },
299
+ {
300
+ "epoch": 0.056622851365015166,
301
+ "grad_norm": 0.4296875,
302
+ "learning_rate": 0.0007,
303
+ "loss": 2.86330623626709,
304
+ "step": 840
305
+ },
306
+ {
307
+ "epoch": 0.057971014492753624,
308
+ "grad_norm": 0.388671875,
309
+ "learning_rate": 0.0007,
310
+ "loss": 2.81943302154541,
311
+ "step": 860
312
+ },
313
+ {
314
+ "epoch": 0.05931917762049208,
315
+ "grad_norm": 0.384765625,
316
+ "learning_rate": 0.0007,
317
+ "loss": 2.7944929122924806,
318
+ "step": 880
319
+ },
320
+ {
321
+ "epoch": 0.06066734074823053,
322
+ "grad_norm": 0.40234375,
323
+ "learning_rate": 0.0007,
324
+ "loss": 2.779962730407715,
325
+ "step": 900
326
+ },
327
+ {
328
+ "epoch": 0.06201550387596899,
329
+ "grad_norm": 0.416015625,
330
+ "learning_rate": 0.0007,
331
+ "loss": 2.751751708984375,
332
+ "step": 920
333
+ },
334
+ {
335
+ "epoch": 0.06336366700370745,
336
+ "grad_norm": 0.447265625,
337
+ "learning_rate": 0.0007,
338
+ "loss": 2.7392576217651365,
339
+ "step": 940
340
+ },
341
+ {
342
+ "epoch": 0.06471183013144591,
343
+ "grad_norm": 0.44140625,
344
+ "learning_rate": 0.0007,
345
+ "loss": 2.7111114501953124,
346
+ "step": 960
347
+ },
348
+ {
349
+ "epoch": 0.06605999325918437,
350
+ "grad_norm": 0.39453125,
351
+ "learning_rate": 0.0007,
352
+ "loss": 2.692020797729492,
353
+ "step": 980
354
+ },
355
+ {
356
+ "epoch": 0.06740815638692282,
357
+ "grad_norm": 0.36328125,
358
+ "learning_rate": 0.0007,
359
+ "loss": 2.691050910949707,
360
+ "step": 1000
361
+ },
362
+ {
363
+ "epoch": 0.06875631951466127,
364
+ "grad_norm": 0.37109375,
365
+ "learning_rate": 0.0007,
366
+ "loss": 2.6589706420898436,
367
+ "step": 1020
368
+ },
369
+ {
370
+ "epoch": 0.07010448264239973,
371
+ "grad_norm": 0.40625,
372
+ "learning_rate": 0.0007,
373
+ "loss": 2.643506622314453,
374
+ "step": 1040
375
+ },
376
+ {
377
+ "epoch": 0.07145264577013818,
378
+ "grad_norm": 0.390625,
379
+ "learning_rate": 0.0007,
380
+ "loss": 2.629362869262695,
381
+ "step": 1060
382
+ },
383
+ {
384
+ "epoch": 0.07280080889787664,
385
+ "grad_norm": 0.37890625,
386
+ "learning_rate": 0.0007,
387
+ "loss": 2.606895637512207,
388
+ "step": 1080
389
+ },
390
+ {
391
+ "epoch": 0.0741489720256151,
392
+ "grad_norm": 0.412109375,
393
+ "learning_rate": 0.0007,
394
+ "loss": 2.6003129959106444,
395
+ "step": 1100
396
+ },
397
+ {
398
+ "epoch": 0.07549713515335356,
399
+ "grad_norm": 0.384765625,
400
+ "learning_rate": 0.0007,
401
+ "loss": 2.565888786315918,
402
+ "step": 1120
403
+ },
404
+ {
405
+ "epoch": 0.07684529828109202,
406
+ "grad_norm": 0.41796875,
407
+ "learning_rate": 0.0007,
408
+ "loss": 2.55434627532959,
409
+ "step": 1140
410
+ },
411
+ {
412
+ "epoch": 0.07819346140883048,
413
+ "grad_norm": 0.37890625,
414
+ "learning_rate": 0.0007,
415
+ "loss": 2.5499151229858397,
416
+ "step": 1160
417
+ },
418
+ {
419
+ "epoch": 0.07954162453656892,
420
+ "grad_norm": 0.38671875,
421
+ "learning_rate": 0.0007,
422
+ "loss": 2.5433088302612306,
423
+ "step": 1180
424
+ },
425
+ {
426
+ "epoch": 0.08088978766430738,
427
+ "grad_norm": 0.45703125,
428
+ "learning_rate": 0.0007,
429
+ "loss": 2.533991050720215,
430
+ "step": 1200
431
+ },
432
+ {
433
+ "epoch": 0.08223795079204584,
434
+ "grad_norm": 0.359375,
435
+ "learning_rate": 0.0007,
436
+ "loss": 2.5188695907592775,
437
+ "step": 1220
438
+ },
439
+ {
440
+ "epoch": 0.0835861139197843,
441
+ "grad_norm": 0.36328125,
442
+ "learning_rate": 0.0007,
443
+ "loss": 2.504729080200195,
444
+ "step": 1240
445
+ },
446
+ {
447
+ "epoch": 0.08493427704752275,
448
+ "grad_norm": 0.3671875,
449
+ "learning_rate": 0.0007,
450
+ "loss": 2.4896400451660154,
451
+ "step": 1260
452
+ },
453
+ {
454
+ "epoch": 0.08628244017526121,
455
+ "grad_norm": 0.369140625,
456
+ "learning_rate": 0.0007,
457
+ "loss": 2.4696292877197266,
458
+ "step": 1280
459
+ },
460
+ {
461
+ "epoch": 0.08763060330299967,
462
+ "grad_norm": 0.369140625,
463
+ "learning_rate": 0.0007,
464
+ "loss": 2.4618251800537108,
465
+ "step": 1300
466
+ },
467
+ {
468
+ "epoch": 0.08897876643073811,
469
+ "grad_norm": 0.361328125,
470
+ "learning_rate": 0.0007,
471
+ "loss": 2.454075813293457,
472
+ "step": 1320
473
+ },
474
+ {
475
+ "epoch": 0.09032692955847657,
476
+ "grad_norm": 0.365234375,
477
+ "learning_rate": 0.0007,
478
+ "loss": 2.441549301147461,
479
+ "step": 1340
480
+ },
481
+ {
482
+ "epoch": 0.09167509268621503,
483
+ "grad_norm": 0.408203125,
484
+ "learning_rate": 0.0007,
485
+ "loss": 2.440728950500488,
486
+ "step": 1360
487
+ },
488
+ {
489
+ "epoch": 0.09302325581395349,
490
+ "grad_norm": 0.369140625,
491
+ "learning_rate": 0.0007,
492
+ "loss": 2.417755889892578,
493
+ "step": 1380
494
+ },
495
+ {
496
+ "epoch": 0.09437141894169195,
497
+ "grad_norm": 0.34765625,
498
+ "learning_rate": 0.0007,
499
+ "loss": 2.408277893066406,
500
+ "step": 1400
501
+ },
502
+ {
503
+ "epoch": 0.0957195820694304,
504
+ "grad_norm": 0.380859375,
505
+ "learning_rate": 0.0007,
506
+ "loss": 2.4170982360839846,
507
+ "step": 1420
508
+ },
509
+ {
510
+ "epoch": 0.09706774519716886,
511
+ "grad_norm": 0.400390625,
512
+ "learning_rate": 0.0007,
513
+ "loss": 2.397001838684082,
514
+ "step": 1440
515
+ },
516
+ {
517
+ "epoch": 0.09841590832490732,
518
+ "grad_norm": 0.380859375,
519
+ "learning_rate": 0.0007,
520
+ "loss": 2.3847665786743164,
521
+ "step": 1460
522
+ },
523
+ {
524
+ "epoch": 0.09976407145264576,
525
+ "grad_norm": 0.357421875,
526
+ "learning_rate": 0.0007,
527
+ "loss": 2.3689859390258787,
528
+ "step": 1480
529
+ },
530
+ {
531
+ "epoch": 0.10111223458038422,
532
+ "grad_norm": 0.365234375,
533
+ "learning_rate": 0.0007,
534
+ "loss": 2.3572067260742187,
535
+ "step": 1500
536
+ },
537
+ {
538
+ "epoch": 0.10246039770812268,
539
+ "grad_norm": 0.359375,
540
+ "learning_rate": 0.0007,
541
+ "loss": 2.3587852478027345,
542
+ "step": 1520
543
+ },
544
+ {
545
+ "epoch": 0.10380856083586114,
546
+ "grad_norm": 0.3984375,
547
+ "learning_rate": 0.0007,
548
+ "loss": 2.3474098205566407,
549
+ "step": 1540
550
+ },
551
+ {
552
+ "epoch": 0.1051567239635996,
553
+ "grad_norm": 0.369140625,
554
+ "learning_rate": 0.0007,
555
+ "loss": 2.336527633666992,
556
+ "step": 1560
557
+ },
558
+ {
559
+ "epoch": 0.10650488709133805,
560
+ "grad_norm": 0.34765625,
561
+ "learning_rate": 0.0007,
562
+ "loss": 2.354816436767578,
563
+ "step": 1580
564
+ },
565
+ {
566
+ "epoch": 0.10785305021907651,
567
+ "grad_norm": 0.40625,
568
+ "learning_rate": 0.0007,
569
+ "loss": 2.3224559783935548,
570
+ "step": 1600
571
+ },
572
+ {
573
+ "epoch": 0.10920121334681497,
574
+ "grad_norm": 0.3515625,
575
+ "learning_rate": 0.0007,
576
+ "loss": 2.320570945739746,
577
+ "step": 1620
578
+ },
579
+ {
580
+ "epoch": 0.11054937647455342,
581
+ "grad_norm": 0.3515625,
582
+ "learning_rate": 0.0007,
583
+ "loss": 2.331516075134277,
584
+ "step": 1640
585
+ },
586
+ {
587
+ "epoch": 0.11189753960229187,
588
+ "grad_norm": 0.326171875,
589
+ "learning_rate": 0.0007,
590
+ "loss": 2.3128894805908202,
591
+ "step": 1660
592
+ },
593
+ {
594
+ "epoch": 0.11324570273003033,
595
+ "grad_norm": 0.341796875,
596
+ "learning_rate": 0.0007,
597
+ "loss": 2.2964305877685547,
598
+ "step": 1680
599
+ },
600
+ {
601
+ "epoch": 0.11459386585776879,
602
+ "grad_norm": 0.376953125,
603
+ "learning_rate": 0.0007,
604
+ "loss": 2.3081363677978515,
605
+ "step": 1700
606
+ },
607
+ {
608
+ "epoch": 0.11594202898550725,
609
+ "grad_norm": 0.375,
610
+ "learning_rate": 0.0007,
611
+ "loss": 2.2839227676391602,
612
+ "step": 1720
613
+ },
614
+ {
615
+ "epoch": 0.1172901921132457,
616
+ "grad_norm": 0.333984375,
617
+ "learning_rate": 0.0007,
618
+ "loss": 2.2729005813598633,
619
+ "step": 1740
620
+ },
621
+ {
622
+ "epoch": 0.11863835524098416,
623
+ "grad_norm": 0.34375,
624
+ "learning_rate": 0.0007,
625
+ "loss": 2.2774417877197264,
626
+ "step": 1760
627
+ },
628
+ {
629
+ "epoch": 0.11998651836872262,
630
+ "grad_norm": 0.390625,
631
+ "learning_rate": 0.0007,
632
+ "loss": 2.290260887145996,
633
+ "step": 1780
634
+ },
635
+ {
636
+ "epoch": 0.12133468149646107,
637
+ "grad_norm": 0.3515625,
638
+ "learning_rate": 0.0007,
639
+ "loss": 2.286345672607422,
640
+ "step": 1800
641
+ },
642
+ {
643
+ "epoch": 0.12268284462419952,
644
+ "grad_norm": 0.396484375,
645
+ "learning_rate": 0.0007,
646
+ "loss": 2.278194236755371,
647
+ "step": 1820
648
+ },
649
+ {
650
+ "epoch": 0.12403100775193798,
651
+ "grad_norm": 0.345703125,
652
+ "learning_rate": 0.0007,
653
+ "loss": 2.245047760009766,
654
+ "step": 1840
655
+ },
656
+ {
657
+ "epoch": 0.12537917087967643,
658
+ "grad_norm": 0.369140625,
659
+ "learning_rate": 0.0007,
660
+ "loss": 2.2449272155761717,
661
+ "step": 1860
662
+ },
663
+ {
664
+ "epoch": 0.1267273340074149,
665
+ "grad_norm": 0.32421875,
666
+ "learning_rate": 0.0007,
667
+ "loss": 2.236011505126953,
668
+ "step": 1880
669
+ },
670
+ {
671
+ "epoch": 0.12807549713515334,
672
+ "grad_norm": 0.353515625,
673
+ "learning_rate": 0.0007,
674
+ "loss": 2.2427194595336912,
675
+ "step": 1900
676
+ },
677
+ {
678
+ "epoch": 0.12942366026289182,
679
+ "grad_norm": 0.341796875,
680
+ "learning_rate": 0.0007,
681
+ "loss": 2.220064163208008,
682
+ "step": 1920
683
+ },
684
+ {
685
+ "epoch": 0.13077182339063026,
686
+ "grad_norm": 0.349609375,
687
+ "learning_rate": 0.0007,
688
+ "loss": 2.230933952331543,
689
+ "step": 1940
690
+ },
691
+ {
692
+ "epoch": 0.13211998651836873,
693
+ "grad_norm": 0.345703125,
694
+ "learning_rate": 0.0007,
695
+ "loss": 2.2043487548828127,
696
+ "step": 1960
697
+ },
698
+ {
699
+ "epoch": 0.13346814964610718,
700
+ "grad_norm": 0.35546875,
701
+ "learning_rate": 0.0007,
702
+ "loss": 2.226978874206543,
703
+ "step": 1980
704
+ },
705
+ {
706
+ "epoch": 0.13481631277384565,
707
+ "grad_norm": 0.341796875,
708
+ "learning_rate": 0.0007,
709
+ "loss": 2.201649284362793,
710
+ "step": 2000
711
+ }
712
+ ],
713
+ "logging_steps": 20,
714
+ "max_steps": 2500,
715
+ "num_input_tokens_seen": 0,
716
+ "num_train_epochs": 1,
717
+ "save_steps": 100,
718
+ "stateful_callbacks": {
719
+ "TrainerControl": {
720
+ "args": {
721
+ "should_epoch_stop": false,
722
+ "should_evaluate": false,
723
+ "should_log": false,
724
+ "should_save": true,
725
+ "should_training_stop": false
726
+ },
727
+ "attributes": {}
728
+ }
729
+ },
730
+ "total_flos": 6452567605248000.0,
731
+ "train_batch_size": 64,
732
+ "trial_name": null,
733
+ "trial_params": null
734
+ }
zain/Activation/out/glu-situglu-100L_run/checkpoint-2000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:26b99898fe48ff0e75b1f0139845f22400023db32e87c8fca3404fda1b772936
3
+ size 4920
zain/Activation/out/glu-situglu-100L_run/checkpoint-2100/config.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation": "situglu",
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": 100,
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-situglu-100L_run/checkpoint-2100/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6bd96bfc4717dda29de4e37e9e2ba42ae6e70272e437d3e83332843310e89b19
3
+ size 33967272
zain/Activation/out/glu-situglu-100L_run/checkpoint-2100/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a2bc9e9cfcc3700ba4e6112363eb84a8e2df3d19ec83b391cf0fe114f9bc5378
3
+ size 68504996
zain/Activation/out/glu-situglu-100L_run/checkpoint-2100/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-situglu-100L_run/checkpoint-2100/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:985eadbe03d32df01f608a67ee99d09997f8b557ed94b6768e8445a622bb3975
3
+ size 1064
zain/Activation/out/glu-situglu-100L_run/checkpoint-2100/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
zain/Activation/out/glu-situglu-100L_run/checkpoint-2100/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-situglu-100L_run/checkpoint-2100/trainer_state.json ADDED
@@ -0,0 +1,769 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 0.14155712841253792,
6
+ "eval_steps": 2498,
7
+ "global_step": 2100,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.0013481631277384564,
14
+ "grad_norm": 2.078125,
15
+ "learning_rate": 2.66e-05,
16
+ "loss": 8.299653625488281,
17
+ "step": 20
18
+ },
19
+ {
20
+ "epoch": 0.002696326255476913,
21
+ "grad_norm": 1.3515625,
22
+ "learning_rate": 5.46e-05,
23
+ "loss": 8.04769058227539,
24
+ "step": 40
25
+ },
26
+ {
27
+ "epoch": 0.004044489383215369,
28
+ "grad_norm": 1.2734375,
29
+ "learning_rate": 8.259999999999999e-05,
30
+ "loss": 7.8379150390625,
31
+ "step": 60
32
+ },
33
+ {
34
+ "epoch": 0.005392652510953826,
35
+ "grad_norm": 1.265625,
36
+ "learning_rate": 0.0001106,
37
+ "loss": 7.572837066650391,
38
+ "step": 80
39
+ },
40
+ {
41
+ "epoch": 0.006740815638692282,
42
+ "grad_norm": 1.1953125,
43
+ "learning_rate": 0.0001386,
44
+ "loss": 7.233182525634765,
45
+ "step": 100
46
+ },
47
+ {
48
+ "epoch": 0.008088978766430738,
49
+ "grad_norm": 1.1328125,
50
+ "learning_rate": 0.00016659999999999998,
51
+ "loss": 6.860689544677735,
52
+ "step": 120
53
+ },
54
+ {
55
+ "epoch": 0.009437141894169195,
56
+ "grad_norm": 0.94921875,
57
+ "learning_rate": 0.00019460000000000001,
58
+ "loss": 6.517751312255859,
59
+ "step": 140
60
+ },
61
+ {
62
+ "epoch": 0.010785305021907651,
63
+ "grad_norm": 0.8046875,
64
+ "learning_rate": 0.0002226,
65
+ "loss": 6.239968109130859,
66
+ "step": 160
67
+ },
68
+ {
69
+ "epoch": 0.012133468149646108,
70
+ "grad_norm": 0.6640625,
71
+ "learning_rate": 0.00025059999999999997,
72
+ "loss": 6.041232299804688,
73
+ "step": 180
74
+ },
75
+ {
76
+ "epoch": 0.013481631277384564,
77
+ "grad_norm": 0.5234375,
78
+ "learning_rate": 0.0002786,
79
+ "loss": 5.863973999023438,
80
+ "step": 200
81
+ },
82
+ {
83
+ "epoch": 0.01482979440512302,
84
+ "grad_norm": 1.375,
85
+ "learning_rate": 0.00030659999999999997,
86
+ "loss": 5.692941665649414,
87
+ "step": 220
88
+ },
89
+ {
90
+ "epoch": 0.016177957532861477,
91
+ "grad_norm": 0.875,
92
+ "learning_rate": 0.0003346,
93
+ "loss": 5.484311294555664,
94
+ "step": 240
95
+ },
96
+ {
97
+ "epoch": 0.01752612066059993,
98
+ "grad_norm": 1.25,
99
+ "learning_rate": 0.00036260000000000003,
100
+ "loss": 5.25721549987793,
101
+ "step": 260
102
+ },
103
+ {
104
+ "epoch": 0.01887428378833839,
105
+ "grad_norm": 0.8125,
106
+ "learning_rate": 0.0003906,
107
+ "loss": 5.028667831420899,
108
+ "step": 280
109
+ },
110
+ {
111
+ "epoch": 0.020222446916076844,
112
+ "grad_norm": 0.75390625,
113
+ "learning_rate": 0.0004186,
114
+ "loss": 4.791546249389649,
115
+ "step": 300
116
+ },
117
+ {
118
+ "epoch": 0.021570610043815303,
119
+ "grad_norm": 0.875,
120
+ "learning_rate": 0.0004466,
121
+ "loss": 4.598950958251953,
122
+ "step": 320
123
+ },
124
+ {
125
+ "epoch": 0.022918773171553757,
126
+ "grad_norm": 0.8125,
127
+ "learning_rate": 0.00047460000000000004,
128
+ "loss": 4.41942024230957,
129
+ "step": 340
130
+ },
131
+ {
132
+ "epoch": 0.024266936299292215,
133
+ "grad_norm": 0.890625,
134
+ "learning_rate": 0.0005026,
135
+ "loss": 4.27253303527832,
136
+ "step": 360
137
+ },
138
+ {
139
+ "epoch": 0.02561509942703067,
140
+ "grad_norm": 1.2734375,
141
+ "learning_rate": 0.0005306,
142
+ "loss": 4.1283222198486325,
143
+ "step": 380
144
+ },
145
+ {
146
+ "epoch": 0.026963262554769128,
147
+ "grad_norm": 0.54296875,
148
+ "learning_rate": 0.0005586,
149
+ "loss": 3.9935020446777343,
150
+ "step": 400
151
+ },
152
+ {
153
+ "epoch": 0.028311425682507583,
154
+ "grad_norm": 0.546875,
155
+ "learning_rate": 0.0005866,
156
+ "loss": 3.885255813598633,
157
+ "step": 420
158
+ },
159
+ {
160
+ "epoch": 0.02965958881024604,
161
+ "grad_norm": 0.7890625,
162
+ "learning_rate": 0.0006146,
163
+ "loss": 3.821950912475586,
164
+ "step": 440
165
+ },
166
+ {
167
+ "epoch": 0.031007751937984496,
168
+ "grad_norm": 0.52734375,
169
+ "learning_rate": 0.0006426,
170
+ "loss": 3.7068336486816404,
171
+ "step": 460
172
+ },
173
+ {
174
+ "epoch": 0.032355915065722954,
175
+ "grad_norm": 0.6328125,
176
+ "learning_rate": 0.0006705999999999999,
177
+ "loss": 3.6370723724365233,
178
+ "step": 480
179
+ },
180
+ {
181
+ "epoch": 0.03370407819346141,
182
+ "grad_norm": 0.5625,
183
+ "learning_rate": 0.0006986,
184
+ "loss": 3.5550304412841798,
185
+ "step": 500
186
+ },
187
+ {
188
+ "epoch": 0.03505224132119986,
189
+ "grad_norm": 0.47265625,
190
+ "learning_rate": 0.0007,
191
+ "loss": 3.4874305725097656,
192
+ "step": 520
193
+ },
194
+ {
195
+ "epoch": 0.03640040444893832,
196
+ "grad_norm": 0.6328125,
197
+ "learning_rate": 0.0007,
198
+ "loss": 3.414617919921875,
199
+ "step": 540
200
+ },
201
+ {
202
+ "epoch": 0.03774856757667678,
203
+ "grad_norm": 0.478515625,
204
+ "learning_rate": 0.0007,
205
+ "loss": 3.3621952056884767,
206
+ "step": 560
207
+ },
208
+ {
209
+ "epoch": 0.03909673070441524,
210
+ "grad_norm": 0.478515625,
211
+ "learning_rate": 0.0007,
212
+ "loss": 3.301148223876953,
213
+ "step": 580
214
+ },
215
+ {
216
+ "epoch": 0.04044489383215369,
217
+ "grad_norm": 0.44140625,
218
+ "learning_rate": 0.0007,
219
+ "loss": 3.266124725341797,
220
+ "step": 600
221
+ },
222
+ {
223
+ "epoch": 0.04179305695989215,
224
+ "grad_norm": 0.58203125,
225
+ "learning_rate": 0.0007,
226
+ "loss": 3.1991281509399414,
227
+ "step": 620
228
+ },
229
+ {
230
+ "epoch": 0.043141220087630605,
231
+ "grad_norm": 0.51953125,
232
+ "learning_rate": 0.0007,
233
+ "loss": 3.175183868408203,
234
+ "step": 640
235
+ },
236
+ {
237
+ "epoch": 0.044489383215369056,
238
+ "grad_norm": 0.443359375,
239
+ "learning_rate": 0.0007,
240
+ "loss": 3.1185443878173826,
241
+ "step": 660
242
+ },
243
+ {
244
+ "epoch": 0.045837546343107514,
245
+ "grad_norm": 0.447265625,
246
+ "learning_rate": 0.0007,
247
+ "loss": 3.091288757324219,
248
+ "step": 680
249
+ },
250
+ {
251
+ "epoch": 0.04718570947084597,
252
+ "grad_norm": 0.400390625,
253
+ "learning_rate": 0.0007,
254
+ "loss": 3.0479175567626955,
255
+ "step": 700
256
+ },
257
+ {
258
+ "epoch": 0.04853387259858443,
259
+ "grad_norm": 0.39453125,
260
+ "learning_rate": 0.0007,
261
+ "loss": 3.012004280090332,
262
+ "step": 720
263
+ },
264
+ {
265
+ "epoch": 0.04988203572632288,
266
+ "grad_norm": 0.40625,
267
+ "learning_rate": 0.0007,
268
+ "loss": 2.973306655883789,
269
+ "step": 740
270
+ },
271
+ {
272
+ "epoch": 0.05123019885406134,
273
+ "grad_norm": 0.41015625,
274
+ "learning_rate": 0.0007,
275
+ "loss": 2.9569366455078123,
276
+ "step": 760
277
+ },
278
+ {
279
+ "epoch": 0.0525783619817998,
280
+ "grad_norm": 0.427734375,
281
+ "learning_rate": 0.0007,
282
+ "loss": 2.959245491027832,
283
+ "step": 780
284
+ },
285
+ {
286
+ "epoch": 0.053926525109538256,
287
+ "grad_norm": 0.4296875,
288
+ "learning_rate": 0.0007,
289
+ "loss": 2.9073596954345704,
290
+ "step": 800
291
+ },
292
+ {
293
+ "epoch": 0.05527468823727671,
294
+ "grad_norm": 0.435546875,
295
+ "learning_rate": 0.0007,
296
+ "loss": 2.894051933288574,
297
+ "step": 820
298
+ },
299
+ {
300
+ "epoch": 0.056622851365015166,
301
+ "grad_norm": 0.4296875,
302
+ "learning_rate": 0.0007,
303
+ "loss": 2.86330623626709,
304
+ "step": 840
305
+ },
306
+ {
307
+ "epoch": 0.057971014492753624,
308
+ "grad_norm": 0.388671875,
309
+ "learning_rate": 0.0007,
310
+ "loss": 2.81943302154541,
311
+ "step": 860
312
+ },
313
+ {
314
+ "epoch": 0.05931917762049208,
315
+ "grad_norm": 0.384765625,
316
+ "learning_rate": 0.0007,
317
+ "loss": 2.7944929122924806,
318
+ "step": 880
319
+ },
320
+ {
321
+ "epoch": 0.06066734074823053,
322
+ "grad_norm": 0.40234375,
323
+ "learning_rate": 0.0007,
324
+ "loss": 2.779962730407715,
325
+ "step": 900
326
+ },
327
+ {
328
+ "epoch": 0.06201550387596899,
329
+ "grad_norm": 0.416015625,
330
+ "learning_rate": 0.0007,
331
+ "loss": 2.751751708984375,
332
+ "step": 920
333
+ },
334
+ {
335
+ "epoch": 0.06336366700370745,
336
+ "grad_norm": 0.447265625,
337
+ "learning_rate": 0.0007,
338
+ "loss": 2.7392576217651365,
339
+ "step": 940
340
+ },
341
+ {
342
+ "epoch": 0.06471183013144591,
343
+ "grad_norm": 0.44140625,
344
+ "learning_rate": 0.0007,
345
+ "loss": 2.7111114501953124,
346
+ "step": 960
347
+ },
348
+ {
349
+ "epoch": 0.06605999325918437,
350
+ "grad_norm": 0.39453125,
351
+ "learning_rate": 0.0007,
352
+ "loss": 2.692020797729492,
353
+ "step": 980
354
+ },
355
+ {
356
+ "epoch": 0.06740815638692282,
357
+ "grad_norm": 0.36328125,
358
+ "learning_rate": 0.0007,
359
+ "loss": 2.691050910949707,
360
+ "step": 1000
361
+ },
362
+ {
363
+ "epoch": 0.06875631951466127,
364
+ "grad_norm": 0.37109375,
365
+ "learning_rate": 0.0007,
366
+ "loss": 2.6589706420898436,
367
+ "step": 1020
368
+ },
369
+ {
370
+ "epoch": 0.07010448264239973,
371
+ "grad_norm": 0.40625,
372
+ "learning_rate": 0.0007,
373
+ "loss": 2.643506622314453,
374
+ "step": 1040
375
+ },
376
+ {
377
+ "epoch": 0.07145264577013818,
378
+ "grad_norm": 0.390625,
379
+ "learning_rate": 0.0007,
380
+ "loss": 2.629362869262695,
381
+ "step": 1060
382
+ },
383
+ {
384
+ "epoch": 0.07280080889787664,
385
+ "grad_norm": 0.37890625,
386
+ "learning_rate": 0.0007,
387
+ "loss": 2.606895637512207,
388
+ "step": 1080
389
+ },
390
+ {
391
+ "epoch": 0.0741489720256151,
392
+ "grad_norm": 0.412109375,
393
+ "learning_rate": 0.0007,
394
+ "loss": 2.6003129959106444,
395
+ "step": 1100
396
+ },
397
+ {
398
+ "epoch": 0.07549713515335356,
399
+ "grad_norm": 0.384765625,
400
+ "learning_rate": 0.0007,
401
+ "loss": 2.565888786315918,
402
+ "step": 1120
403
+ },
404
+ {
405
+ "epoch": 0.07684529828109202,
406
+ "grad_norm": 0.41796875,
407
+ "learning_rate": 0.0007,
408
+ "loss": 2.55434627532959,
409
+ "step": 1140
410
+ },
411
+ {
412
+ "epoch": 0.07819346140883048,
413
+ "grad_norm": 0.37890625,
414
+ "learning_rate": 0.0007,
415
+ "loss": 2.5499151229858397,
416
+ "step": 1160
417
+ },
418
+ {
419
+ "epoch": 0.07954162453656892,
420
+ "grad_norm": 0.38671875,
421
+ "learning_rate": 0.0007,
422
+ "loss": 2.5433088302612306,
423
+ "step": 1180
424
+ },
425
+ {
426
+ "epoch": 0.08088978766430738,
427
+ "grad_norm": 0.45703125,
428
+ "learning_rate": 0.0007,
429
+ "loss": 2.533991050720215,
430
+ "step": 1200
431
+ },
432
+ {
433
+ "epoch": 0.08223795079204584,
434
+ "grad_norm": 0.359375,
435
+ "learning_rate": 0.0007,
436
+ "loss": 2.5188695907592775,
437
+ "step": 1220
438
+ },
439
+ {
440
+ "epoch": 0.0835861139197843,
441
+ "grad_norm": 0.36328125,
442
+ "learning_rate": 0.0007,
443
+ "loss": 2.504729080200195,
444
+ "step": 1240
445
+ },
446
+ {
447
+ "epoch": 0.08493427704752275,
448
+ "grad_norm": 0.3671875,
449
+ "learning_rate": 0.0007,
450
+ "loss": 2.4896400451660154,
451
+ "step": 1260
452
+ },
453
+ {
454
+ "epoch": 0.08628244017526121,
455
+ "grad_norm": 0.369140625,
456
+ "learning_rate": 0.0007,
457
+ "loss": 2.4696292877197266,
458
+ "step": 1280
459
+ },
460
+ {
461
+ "epoch": 0.08763060330299967,
462
+ "grad_norm": 0.369140625,
463
+ "learning_rate": 0.0007,
464
+ "loss": 2.4618251800537108,
465
+ "step": 1300
466
+ },
467
+ {
468
+ "epoch": 0.08897876643073811,
469
+ "grad_norm": 0.361328125,
470
+ "learning_rate": 0.0007,
471
+ "loss": 2.454075813293457,
472
+ "step": 1320
473
+ },
474
+ {
475
+ "epoch": 0.09032692955847657,
476
+ "grad_norm": 0.365234375,
477
+ "learning_rate": 0.0007,
478
+ "loss": 2.441549301147461,
479
+ "step": 1340
480
+ },
481
+ {
482
+ "epoch": 0.09167509268621503,
483
+ "grad_norm": 0.408203125,
484
+ "learning_rate": 0.0007,
485
+ "loss": 2.440728950500488,
486
+ "step": 1360
487
+ },
488
+ {
489
+ "epoch": 0.09302325581395349,
490
+ "grad_norm": 0.369140625,
491
+ "learning_rate": 0.0007,
492
+ "loss": 2.417755889892578,
493
+ "step": 1380
494
+ },
495
+ {
496
+ "epoch": 0.09437141894169195,
497
+ "grad_norm": 0.34765625,
498
+ "learning_rate": 0.0007,
499
+ "loss": 2.408277893066406,
500
+ "step": 1400
501
+ },
502
+ {
503
+ "epoch": 0.0957195820694304,
504
+ "grad_norm": 0.380859375,
505
+ "learning_rate": 0.0007,
506
+ "loss": 2.4170982360839846,
507
+ "step": 1420
508
+ },
509
+ {
510
+ "epoch": 0.09706774519716886,
511
+ "grad_norm": 0.400390625,
512
+ "learning_rate": 0.0007,
513
+ "loss": 2.397001838684082,
514
+ "step": 1440
515
+ },
516
+ {
517
+ "epoch": 0.09841590832490732,
518
+ "grad_norm": 0.380859375,
519
+ "learning_rate": 0.0007,
520
+ "loss": 2.3847665786743164,
521
+ "step": 1460
522
+ },
523
+ {
524
+ "epoch": 0.09976407145264576,
525
+ "grad_norm": 0.357421875,
526
+ "learning_rate": 0.0007,
527
+ "loss": 2.3689859390258787,
528
+ "step": 1480
529
+ },
530
+ {
531
+ "epoch": 0.10111223458038422,
532
+ "grad_norm": 0.365234375,
533
+ "learning_rate": 0.0007,
534
+ "loss": 2.3572067260742187,
535
+ "step": 1500
536
+ },
537
+ {
538
+ "epoch": 0.10246039770812268,
539
+ "grad_norm": 0.359375,
540
+ "learning_rate": 0.0007,
541
+ "loss": 2.3587852478027345,
542
+ "step": 1520
543
+ },
544
+ {
545
+ "epoch": 0.10380856083586114,
546
+ "grad_norm": 0.3984375,
547
+ "learning_rate": 0.0007,
548
+ "loss": 2.3474098205566407,
549
+ "step": 1540
550
+ },
551
+ {
552
+ "epoch": 0.1051567239635996,
553
+ "grad_norm": 0.369140625,
554
+ "learning_rate": 0.0007,
555
+ "loss": 2.336527633666992,
556
+ "step": 1560
557
+ },
558
+ {
559
+ "epoch": 0.10650488709133805,
560
+ "grad_norm": 0.34765625,
561
+ "learning_rate": 0.0007,
562
+ "loss": 2.354816436767578,
563
+ "step": 1580
564
+ },
565
+ {
566
+ "epoch": 0.10785305021907651,
567
+ "grad_norm": 0.40625,
568
+ "learning_rate": 0.0007,
569
+ "loss": 2.3224559783935548,
570
+ "step": 1600
571
+ },
572
+ {
573
+ "epoch": 0.10920121334681497,
574
+ "grad_norm": 0.3515625,
575
+ "learning_rate": 0.0007,
576
+ "loss": 2.320570945739746,
577
+ "step": 1620
578
+ },
579
+ {
580
+ "epoch": 0.11054937647455342,
581
+ "grad_norm": 0.3515625,
582
+ "learning_rate": 0.0007,
583
+ "loss": 2.331516075134277,
584
+ "step": 1640
585
+ },
586
+ {
587
+ "epoch": 0.11189753960229187,
588
+ "grad_norm": 0.326171875,
589
+ "learning_rate": 0.0007,
590
+ "loss": 2.3128894805908202,
591
+ "step": 1660
592
+ },
593
+ {
594
+ "epoch": 0.11324570273003033,
595
+ "grad_norm": 0.341796875,
596
+ "learning_rate": 0.0007,
597
+ "loss": 2.2964305877685547,
598
+ "step": 1680
599
+ },
600
+ {
601
+ "epoch": 0.11459386585776879,
602
+ "grad_norm": 0.376953125,
603
+ "learning_rate": 0.0007,
604
+ "loss": 2.3081363677978515,
605
+ "step": 1700
606
+ },
607
+ {
608
+ "epoch": 0.11594202898550725,
609
+ "grad_norm": 0.375,
610
+ "learning_rate": 0.0007,
611
+ "loss": 2.2839227676391602,
612
+ "step": 1720
613
+ },
614
+ {
615
+ "epoch": 0.1172901921132457,
616
+ "grad_norm": 0.333984375,
617
+ "learning_rate": 0.0007,
618
+ "loss": 2.2729005813598633,
619
+ "step": 1740
620
+ },
621
+ {
622
+ "epoch": 0.11863835524098416,
623
+ "grad_norm": 0.34375,
624
+ "learning_rate": 0.0007,
625
+ "loss": 2.2774417877197264,
626
+ "step": 1760
627
+ },
628
+ {
629
+ "epoch": 0.11998651836872262,
630
+ "grad_norm": 0.390625,
631
+ "learning_rate": 0.0007,
632
+ "loss": 2.290260887145996,
633
+ "step": 1780
634
+ },
635
+ {
636
+ "epoch": 0.12133468149646107,
637
+ "grad_norm": 0.3515625,
638
+ "learning_rate": 0.0007,
639
+ "loss": 2.286345672607422,
640
+ "step": 1800
641
+ },
642
+ {
643
+ "epoch": 0.12268284462419952,
644
+ "grad_norm": 0.396484375,
645
+ "learning_rate": 0.0007,
646
+ "loss": 2.278194236755371,
647
+ "step": 1820
648
+ },
649
+ {
650
+ "epoch": 0.12403100775193798,
651
+ "grad_norm": 0.345703125,
652
+ "learning_rate": 0.0007,
653
+ "loss": 2.245047760009766,
654
+ "step": 1840
655
+ },
656
+ {
657
+ "epoch": 0.12537917087967643,
658
+ "grad_norm": 0.369140625,
659
+ "learning_rate": 0.0007,
660
+ "loss": 2.2449272155761717,
661
+ "step": 1860
662
+ },
663
+ {
664
+ "epoch": 0.1267273340074149,
665
+ "grad_norm": 0.32421875,
666
+ "learning_rate": 0.0007,
667
+ "loss": 2.236011505126953,
668
+ "step": 1880
669
+ },
670
+ {
671
+ "epoch": 0.12807549713515334,
672
+ "grad_norm": 0.353515625,
673
+ "learning_rate": 0.0007,
674
+ "loss": 2.2427194595336912,
675
+ "step": 1900
676
+ },
677
+ {
678
+ "epoch": 0.12942366026289182,
679
+ "grad_norm": 0.341796875,
680
+ "learning_rate": 0.0007,
681
+ "loss": 2.220064163208008,
682
+ "step": 1920
683
+ },
684
+ {
685
+ "epoch": 0.13077182339063026,
686
+ "grad_norm": 0.349609375,
687
+ "learning_rate": 0.0007,
688
+ "loss": 2.230933952331543,
689
+ "step": 1940
690
+ },
691
+ {
692
+ "epoch": 0.13211998651836873,
693
+ "grad_norm": 0.345703125,
694
+ "learning_rate": 0.0007,
695
+ "loss": 2.2043487548828127,
696
+ "step": 1960
697
+ },
698
+ {
699
+ "epoch": 0.13346814964610718,
700
+ "grad_norm": 0.35546875,
701
+ "learning_rate": 0.0007,
702
+ "loss": 2.226978874206543,
703
+ "step": 1980
704
+ },
705
+ {
706
+ "epoch": 0.13481631277384565,
707
+ "grad_norm": 0.341796875,
708
+ "learning_rate": 0.0007,
709
+ "loss": 2.201649284362793,
710
+ "step": 2000
711
+ },
712
+ {
713
+ "epoch": 0.1361644759015841,
714
+ "grad_norm": 0.3515625,
715
+ "learning_rate": 0.0007,
716
+ "loss": 2.2095453262329103,
717
+ "step": 2020
718
+ },
719
+ {
720
+ "epoch": 0.13751263902932254,
721
+ "grad_norm": 0.376953125,
722
+ "learning_rate": 0.0007,
723
+ "loss": 2.189457893371582,
724
+ "step": 2040
725
+ },
726
+ {
727
+ "epoch": 0.138860802157061,
728
+ "grad_norm": 0.349609375,
729
+ "learning_rate": 0.0007,
730
+ "loss": 2.2083295822143554,
731
+ "step": 2060
732
+ },
733
+ {
734
+ "epoch": 0.14020896528479945,
735
+ "grad_norm": 0.328125,
736
+ "learning_rate": 0.0007,
737
+ "loss": 2.1837480545043944,
738
+ "step": 2080
739
+ },
740
+ {
741
+ "epoch": 0.14155712841253792,
742
+ "grad_norm": 0.330078125,
743
+ "learning_rate": 0.0007,
744
+ "loss": 2.1739730834960938,
745
+ "step": 2100
746
+ }
747
+ ],
748
+ "logging_steps": 20,
749
+ "max_steps": 2500,
750
+ "num_input_tokens_seen": 0,
751
+ "num_train_epochs": 1,
752
+ "save_steps": 100,
753
+ "stateful_callbacks": {
754
+ "TrainerControl": {
755
+ "args": {
756
+ "should_epoch_stop": false,
757
+ "should_evaluate": false,
758
+ "should_log": false,
759
+ "should_save": true,
760
+ "should_training_stop": false
761
+ },
762
+ "attributes": {}
763
+ }
764
+ },
765
+ "total_flos": 6775195985510400.0,
766
+ "train_batch_size": 64,
767
+ "trial_name": null,
768
+ "trial_params": null
769
+ }
zain/Activation/out/glu-situglu-100L_run/checkpoint-2100/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:26b99898fe48ff0e75b1f0139845f22400023db32e87c8fca3404fda1b772936
3
+ size 4920
zain/Activation/out/glu-situglu-100L_run/checkpoint-2200/config.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation": "situglu",
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": 100,
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-situglu-100L_run/checkpoint-2200/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4378f0ac3b317f799a361b8286486ba8377f700e9479e512a69b810a52c5d972
3
+ size 33967272
zain/Activation/out/glu-situglu-100L_run/checkpoint-2200/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:77f6d9fcfb1fb7b5b681796b2dd589d2aed710e240e63b55295e998b3ff1bdb7
3
+ size 68504996
zain/Activation/out/glu-situglu-100L_run/checkpoint-2200/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-situglu-100L_run/checkpoint-2200/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bb38d8fe479c6ce36db4ed357fb0375cafd5d59c32c20acad02585958763071c
3
+ size 1064
zain/Activation/out/glu-situglu-100L_run/checkpoint-2200/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
zain/Activation/out/glu-situglu-100L_run/checkpoint-2200/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-situglu-100L_run/checkpoint-2200/trainer_state.json ADDED
@@ -0,0 +1,804 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 0.1482979440512302,
6
+ "eval_steps": 2498,
7
+ "global_step": 2200,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.0013481631277384564,
14
+ "grad_norm": 2.078125,
15
+ "learning_rate": 2.66e-05,
16
+ "loss": 8.299653625488281,
17
+ "step": 20
18
+ },
19
+ {
20
+ "epoch": 0.002696326255476913,
21
+ "grad_norm": 1.3515625,
22
+ "learning_rate": 5.46e-05,
23
+ "loss": 8.04769058227539,
24
+ "step": 40
25
+ },
26
+ {
27
+ "epoch": 0.004044489383215369,
28
+ "grad_norm": 1.2734375,
29
+ "learning_rate": 8.259999999999999e-05,
30
+ "loss": 7.8379150390625,
31
+ "step": 60
32
+ },
33
+ {
34
+ "epoch": 0.005392652510953826,
35
+ "grad_norm": 1.265625,
36
+ "learning_rate": 0.0001106,
37
+ "loss": 7.572837066650391,
38
+ "step": 80
39
+ },
40
+ {
41
+ "epoch": 0.006740815638692282,
42
+ "grad_norm": 1.1953125,
43
+ "learning_rate": 0.0001386,
44
+ "loss": 7.233182525634765,
45
+ "step": 100
46
+ },
47
+ {
48
+ "epoch": 0.008088978766430738,
49
+ "grad_norm": 1.1328125,
50
+ "learning_rate": 0.00016659999999999998,
51
+ "loss": 6.860689544677735,
52
+ "step": 120
53
+ },
54
+ {
55
+ "epoch": 0.009437141894169195,
56
+ "grad_norm": 0.94921875,
57
+ "learning_rate": 0.00019460000000000001,
58
+ "loss": 6.517751312255859,
59
+ "step": 140
60
+ },
61
+ {
62
+ "epoch": 0.010785305021907651,
63
+ "grad_norm": 0.8046875,
64
+ "learning_rate": 0.0002226,
65
+ "loss": 6.239968109130859,
66
+ "step": 160
67
+ },
68
+ {
69
+ "epoch": 0.012133468149646108,
70
+ "grad_norm": 0.6640625,
71
+ "learning_rate": 0.00025059999999999997,
72
+ "loss": 6.041232299804688,
73
+ "step": 180
74
+ },
75
+ {
76
+ "epoch": 0.013481631277384564,
77
+ "grad_norm": 0.5234375,
78
+ "learning_rate": 0.0002786,
79
+ "loss": 5.863973999023438,
80
+ "step": 200
81
+ },
82
+ {
83
+ "epoch": 0.01482979440512302,
84
+ "grad_norm": 1.375,
85
+ "learning_rate": 0.00030659999999999997,
86
+ "loss": 5.692941665649414,
87
+ "step": 220
88
+ },
89
+ {
90
+ "epoch": 0.016177957532861477,
91
+ "grad_norm": 0.875,
92
+ "learning_rate": 0.0003346,
93
+ "loss": 5.484311294555664,
94
+ "step": 240
95
+ },
96
+ {
97
+ "epoch": 0.01752612066059993,
98
+ "grad_norm": 1.25,
99
+ "learning_rate": 0.00036260000000000003,
100
+ "loss": 5.25721549987793,
101
+ "step": 260
102
+ },
103
+ {
104
+ "epoch": 0.01887428378833839,
105
+ "grad_norm": 0.8125,
106
+ "learning_rate": 0.0003906,
107
+ "loss": 5.028667831420899,
108
+ "step": 280
109
+ },
110
+ {
111
+ "epoch": 0.020222446916076844,
112
+ "grad_norm": 0.75390625,
113
+ "learning_rate": 0.0004186,
114
+ "loss": 4.791546249389649,
115
+ "step": 300
116
+ },
117
+ {
118
+ "epoch": 0.021570610043815303,
119
+ "grad_norm": 0.875,
120
+ "learning_rate": 0.0004466,
121
+ "loss": 4.598950958251953,
122
+ "step": 320
123
+ },
124
+ {
125
+ "epoch": 0.022918773171553757,
126
+ "grad_norm": 0.8125,
127
+ "learning_rate": 0.00047460000000000004,
128
+ "loss": 4.41942024230957,
129
+ "step": 340
130
+ },
131
+ {
132
+ "epoch": 0.024266936299292215,
133
+ "grad_norm": 0.890625,
134
+ "learning_rate": 0.0005026,
135
+ "loss": 4.27253303527832,
136
+ "step": 360
137
+ },
138
+ {
139
+ "epoch": 0.02561509942703067,
140
+ "grad_norm": 1.2734375,
141
+ "learning_rate": 0.0005306,
142
+ "loss": 4.1283222198486325,
143
+ "step": 380
144
+ },
145
+ {
146
+ "epoch": 0.026963262554769128,
147
+ "grad_norm": 0.54296875,
148
+ "learning_rate": 0.0005586,
149
+ "loss": 3.9935020446777343,
150
+ "step": 400
151
+ },
152
+ {
153
+ "epoch": 0.028311425682507583,
154
+ "grad_norm": 0.546875,
155
+ "learning_rate": 0.0005866,
156
+ "loss": 3.885255813598633,
157
+ "step": 420
158
+ },
159
+ {
160
+ "epoch": 0.02965958881024604,
161
+ "grad_norm": 0.7890625,
162
+ "learning_rate": 0.0006146,
163
+ "loss": 3.821950912475586,
164
+ "step": 440
165
+ },
166
+ {
167
+ "epoch": 0.031007751937984496,
168
+ "grad_norm": 0.52734375,
169
+ "learning_rate": 0.0006426,
170
+ "loss": 3.7068336486816404,
171
+ "step": 460
172
+ },
173
+ {
174
+ "epoch": 0.032355915065722954,
175
+ "grad_norm": 0.6328125,
176
+ "learning_rate": 0.0006705999999999999,
177
+ "loss": 3.6370723724365233,
178
+ "step": 480
179
+ },
180
+ {
181
+ "epoch": 0.03370407819346141,
182
+ "grad_norm": 0.5625,
183
+ "learning_rate": 0.0006986,
184
+ "loss": 3.5550304412841798,
185
+ "step": 500
186
+ },
187
+ {
188
+ "epoch": 0.03505224132119986,
189
+ "grad_norm": 0.47265625,
190
+ "learning_rate": 0.0007,
191
+ "loss": 3.4874305725097656,
192
+ "step": 520
193
+ },
194
+ {
195
+ "epoch": 0.03640040444893832,
196
+ "grad_norm": 0.6328125,
197
+ "learning_rate": 0.0007,
198
+ "loss": 3.414617919921875,
199
+ "step": 540
200
+ },
201
+ {
202
+ "epoch": 0.03774856757667678,
203
+ "grad_norm": 0.478515625,
204
+ "learning_rate": 0.0007,
205
+ "loss": 3.3621952056884767,
206
+ "step": 560
207
+ },
208
+ {
209
+ "epoch": 0.03909673070441524,
210
+ "grad_norm": 0.478515625,
211
+ "learning_rate": 0.0007,
212
+ "loss": 3.301148223876953,
213
+ "step": 580
214
+ },
215
+ {
216
+ "epoch": 0.04044489383215369,
217
+ "grad_norm": 0.44140625,
218
+ "learning_rate": 0.0007,
219
+ "loss": 3.266124725341797,
220
+ "step": 600
221
+ },
222
+ {
223
+ "epoch": 0.04179305695989215,
224
+ "grad_norm": 0.58203125,
225
+ "learning_rate": 0.0007,
226
+ "loss": 3.1991281509399414,
227
+ "step": 620
228
+ },
229
+ {
230
+ "epoch": 0.043141220087630605,
231
+ "grad_norm": 0.51953125,
232
+ "learning_rate": 0.0007,
233
+ "loss": 3.175183868408203,
234
+ "step": 640
235
+ },
236
+ {
237
+ "epoch": 0.044489383215369056,
238
+ "grad_norm": 0.443359375,
239
+ "learning_rate": 0.0007,
240
+ "loss": 3.1185443878173826,
241
+ "step": 660
242
+ },
243
+ {
244
+ "epoch": 0.045837546343107514,
245
+ "grad_norm": 0.447265625,
246
+ "learning_rate": 0.0007,
247
+ "loss": 3.091288757324219,
248
+ "step": 680
249
+ },
250
+ {
251
+ "epoch": 0.04718570947084597,
252
+ "grad_norm": 0.400390625,
253
+ "learning_rate": 0.0007,
254
+ "loss": 3.0479175567626955,
255
+ "step": 700
256
+ },
257
+ {
258
+ "epoch": 0.04853387259858443,
259
+ "grad_norm": 0.39453125,
260
+ "learning_rate": 0.0007,
261
+ "loss": 3.012004280090332,
262
+ "step": 720
263
+ },
264
+ {
265
+ "epoch": 0.04988203572632288,
266
+ "grad_norm": 0.40625,
267
+ "learning_rate": 0.0007,
268
+ "loss": 2.973306655883789,
269
+ "step": 740
270
+ },
271
+ {
272
+ "epoch": 0.05123019885406134,
273
+ "grad_norm": 0.41015625,
274
+ "learning_rate": 0.0007,
275
+ "loss": 2.9569366455078123,
276
+ "step": 760
277
+ },
278
+ {
279
+ "epoch": 0.0525783619817998,
280
+ "grad_norm": 0.427734375,
281
+ "learning_rate": 0.0007,
282
+ "loss": 2.959245491027832,
283
+ "step": 780
284
+ },
285
+ {
286
+ "epoch": 0.053926525109538256,
287
+ "grad_norm": 0.4296875,
288
+ "learning_rate": 0.0007,
289
+ "loss": 2.9073596954345704,
290
+ "step": 800
291
+ },
292
+ {
293
+ "epoch": 0.05527468823727671,
294
+ "grad_norm": 0.435546875,
295
+ "learning_rate": 0.0007,
296
+ "loss": 2.894051933288574,
297
+ "step": 820
298
+ },
299
+ {
300
+ "epoch": 0.056622851365015166,
301
+ "grad_norm": 0.4296875,
302
+ "learning_rate": 0.0007,
303
+ "loss": 2.86330623626709,
304
+ "step": 840
305
+ },
306
+ {
307
+ "epoch": 0.057971014492753624,
308
+ "grad_norm": 0.388671875,
309
+ "learning_rate": 0.0007,
310
+ "loss": 2.81943302154541,
311
+ "step": 860
312
+ },
313
+ {
314
+ "epoch": 0.05931917762049208,
315
+ "grad_norm": 0.384765625,
316
+ "learning_rate": 0.0007,
317
+ "loss": 2.7944929122924806,
318
+ "step": 880
319
+ },
320
+ {
321
+ "epoch": 0.06066734074823053,
322
+ "grad_norm": 0.40234375,
323
+ "learning_rate": 0.0007,
324
+ "loss": 2.779962730407715,
325
+ "step": 900
326
+ },
327
+ {
328
+ "epoch": 0.06201550387596899,
329
+ "grad_norm": 0.416015625,
330
+ "learning_rate": 0.0007,
331
+ "loss": 2.751751708984375,
332
+ "step": 920
333
+ },
334
+ {
335
+ "epoch": 0.06336366700370745,
336
+ "grad_norm": 0.447265625,
337
+ "learning_rate": 0.0007,
338
+ "loss": 2.7392576217651365,
339
+ "step": 940
340
+ },
341
+ {
342
+ "epoch": 0.06471183013144591,
343
+ "grad_norm": 0.44140625,
344
+ "learning_rate": 0.0007,
345
+ "loss": 2.7111114501953124,
346
+ "step": 960
347
+ },
348
+ {
349
+ "epoch": 0.06605999325918437,
350
+ "grad_norm": 0.39453125,
351
+ "learning_rate": 0.0007,
352
+ "loss": 2.692020797729492,
353
+ "step": 980
354
+ },
355
+ {
356
+ "epoch": 0.06740815638692282,
357
+ "grad_norm": 0.36328125,
358
+ "learning_rate": 0.0007,
359
+ "loss": 2.691050910949707,
360
+ "step": 1000
361
+ },
362
+ {
363
+ "epoch": 0.06875631951466127,
364
+ "grad_norm": 0.37109375,
365
+ "learning_rate": 0.0007,
366
+ "loss": 2.6589706420898436,
367
+ "step": 1020
368
+ },
369
+ {
370
+ "epoch": 0.07010448264239973,
371
+ "grad_norm": 0.40625,
372
+ "learning_rate": 0.0007,
373
+ "loss": 2.643506622314453,
374
+ "step": 1040
375
+ },
376
+ {
377
+ "epoch": 0.07145264577013818,
378
+ "grad_norm": 0.390625,
379
+ "learning_rate": 0.0007,
380
+ "loss": 2.629362869262695,
381
+ "step": 1060
382
+ },
383
+ {
384
+ "epoch": 0.07280080889787664,
385
+ "grad_norm": 0.37890625,
386
+ "learning_rate": 0.0007,
387
+ "loss": 2.606895637512207,
388
+ "step": 1080
389
+ },
390
+ {
391
+ "epoch": 0.0741489720256151,
392
+ "grad_norm": 0.412109375,
393
+ "learning_rate": 0.0007,
394
+ "loss": 2.6003129959106444,
395
+ "step": 1100
396
+ },
397
+ {
398
+ "epoch": 0.07549713515335356,
399
+ "grad_norm": 0.384765625,
400
+ "learning_rate": 0.0007,
401
+ "loss": 2.565888786315918,
402
+ "step": 1120
403
+ },
404
+ {
405
+ "epoch": 0.07684529828109202,
406
+ "grad_norm": 0.41796875,
407
+ "learning_rate": 0.0007,
408
+ "loss": 2.55434627532959,
409
+ "step": 1140
410
+ },
411
+ {
412
+ "epoch": 0.07819346140883048,
413
+ "grad_norm": 0.37890625,
414
+ "learning_rate": 0.0007,
415
+ "loss": 2.5499151229858397,
416
+ "step": 1160
417
+ },
418
+ {
419
+ "epoch": 0.07954162453656892,
420
+ "grad_norm": 0.38671875,
421
+ "learning_rate": 0.0007,
422
+ "loss": 2.5433088302612306,
423
+ "step": 1180
424
+ },
425
+ {
426
+ "epoch": 0.08088978766430738,
427
+ "grad_norm": 0.45703125,
428
+ "learning_rate": 0.0007,
429
+ "loss": 2.533991050720215,
430
+ "step": 1200
431
+ },
432
+ {
433
+ "epoch": 0.08223795079204584,
434
+ "grad_norm": 0.359375,
435
+ "learning_rate": 0.0007,
436
+ "loss": 2.5188695907592775,
437
+ "step": 1220
438
+ },
439
+ {
440
+ "epoch": 0.0835861139197843,
441
+ "grad_norm": 0.36328125,
442
+ "learning_rate": 0.0007,
443
+ "loss": 2.504729080200195,
444
+ "step": 1240
445
+ },
446
+ {
447
+ "epoch": 0.08493427704752275,
448
+ "grad_norm": 0.3671875,
449
+ "learning_rate": 0.0007,
450
+ "loss": 2.4896400451660154,
451
+ "step": 1260
452
+ },
453
+ {
454
+ "epoch": 0.08628244017526121,
455
+ "grad_norm": 0.369140625,
456
+ "learning_rate": 0.0007,
457
+ "loss": 2.4696292877197266,
458
+ "step": 1280
459
+ },
460
+ {
461
+ "epoch": 0.08763060330299967,
462
+ "grad_norm": 0.369140625,
463
+ "learning_rate": 0.0007,
464
+ "loss": 2.4618251800537108,
465
+ "step": 1300
466
+ },
467
+ {
468
+ "epoch": 0.08897876643073811,
469
+ "grad_norm": 0.361328125,
470
+ "learning_rate": 0.0007,
471
+ "loss": 2.454075813293457,
472
+ "step": 1320
473
+ },
474
+ {
475
+ "epoch": 0.09032692955847657,
476
+ "grad_norm": 0.365234375,
477
+ "learning_rate": 0.0007,
478
+ "loss": 2.441549301147461,
479
+ "step": 1340
480
+ },
481
+ {
482
+ "epoch": 0.09167509268621503,
483
+ "grad_norm": 0.408203125,
484
+ "learning_rate": 0.0007,
485
+ "loss": 2.440728950500488,
486
+ "step": 1360
487
+ },
488
+ {
489
+ "epoch": 0.09302325581395349,
490
+ "grad_norm": 0.369140625,
491
+ "learning_rate": 0.0007,
492
+ "loss": 2.417755889892578,
493
+ "step": 1380
494
+ },
495
+ {
496
+ "epoch": 0.09437141894169195,
497
+ "grad_norm": 0.34765625,
498
+ "learning_rate": 0.0007,
499
+ "loss": 2.408277893066406,
500
+ "step": 1400
501
+ },
502
+ {
503
+ "epoch": 0.0957195820694304,
504
+ "grad_norm": 0.380859375,
505
+ "learning_rate": 0.0007,
506
+ "loss": 2.4170982360839846,
507
+ "step": 1420
508
+ },
509
+ {
510
+ "epoch": 0.09706774519716886,
511
+ "grad_norm": 0.400390625,
512
+ "learning_rate": 0.0007,
513
+ "loss": 2.397001838684082,
514
+ "step": 1440
515
+ },
516
+ {
517
+ "epoch": 0.09841590832490732,
518
+ "grad_norm": 0.380859375,
519
+ "learning_rate": 0.0007,
520
+ "loss": 2.3847665786743164,
521
+ "step": 1460
522
+ },
523
+ {
524
+ "epoch": 0.09976407145264576,
525
+ "grad_norm": 0.357421875,
526
+ "learning_rate": 0.0007,
527
+ "loss": 2.3689859390258787,
528
+ "step": 1480
529
+ },
530
+ {
531
+ "epoch": 0.10111223458038422,
532
+ "grad_norm": 0.365234375,
533
+ "learning_rate": 0.0007,
534
+ "loss": 2.3572067260742187,
535
+ "step": 1500
536
+ },
537
+ {
538
+ "epoch": 0.10246039770812268,
539
+ "grad_norm": 0.359375,
540
+ "learning_rate": 0.0007,
541
+ "loss": 2.3587852478027345,
542
+ "step": 1520
543
+ },
544
+ {
545
+ "epoch": 0.10380856083586114,
546
+ "grad_norm": 0.3984375,
547
+ "learning_rate": 0.0007,
548
+ "loss": 2.3474098205566407,
549
+ "step": 1540
550
+ },
551
+ {
552
+ "epoch": 0.1051567239635996,
553
+ "grad_norm": 0.369140625,
554
+ "learning_rate": 0.0007,
555
+ "loss": 2.336527633666992,
556
+ "step": 1560
557
+ },
558
+ {
559
+ "epoch": 0.10650488709133805,
560
+ "grad_norm": 0.34765625,
561
+ "learning_rate": 0.0007,
562
+ "loss": 2.354816436767578,
563
+ "step": 1580
564
+ },
565
+ {
566
+ "epoch": 0.10785305021907651,
567
+ "grad_norm": 0.40625,
568
+ "learning_rate": 0.0007,
569
+ "loss": 2.3224559783935548,
570
+ "step": 1600
571
+ },
572
+ {
573
+ "epoch": 0.10920121334681497,
574
+ "grad_norm": 0.3515625,
575
+ "learning_rate": 0.0007,
576
+ "loss": 2.320570945739746,
577
+ "step": 1620
578
+ },
579
+ {
580
+ "epoch": 0.11054937647455342,
581
+ "grad_norm": 0.3515625,
582
+ "learning_rate": 0.0007,
583
+ "loss": 2.331516075134277,
584
+ "step": 1640
585
+ },
586
+ {
587
+ "epoch": 0.11189753960229187,
588
+ "grad_norm": 0.326171875,
589
+ "learning_rate": 0.0007,
590
+ "loss": 2.3128894805908202,
591
+ "step": 1660
592
+ },
593
+ {
594
+ "epoch": 0.11324570273003033,
595
+ "grad_norm": 0.341796875,
596
+ "learning_rate": 0.0007,
597
+ "loss": 2.2964305877685547,
598
+ "step": 1680
599
+ },
600
+ {
601
+ "epoch": 0.11459386585776879,
602
+ "grad_norm": 0.376953125,
603
+ "learning_rate": 0.0007,
604
+ "loss": 2.3081363677978515,
605
+ "step": 1700
606
+ },
607
+ {
608
+ "epoch": 0.11594202898550725,
609
+ "grad_norm": 0.375,
610
+ "learning_rate": 0.0007,
611
+ "loss": 2.2839227676391602,
612
+ "step": 1720
613
+ },
614
+ {
615
+ "epoch": 0.1172901921132457,
616
+ "grad_norm": 0.333984375,
617
+ "learning_rate": 0.0007,
618
+ "loss": 2.2729005813598633,
619
+ "step": 1740
620
+ },
621
+ {
622
+ "epoch": 0.11863835524098416,
623
+ "grad_norm": 0.34375,
624
+ "learning_rate": 0.0007,
625
+ "loss": 2.2774417877197264,
626
+ "step": 1760
627
+ },
628
+ {
629
+ "epoch": 0.11998651836872262,
630
+ "grad_norm": 0.390625,
631
+ "learning_rate": 0.0007,
632
+ "loss": 2.290260887145996,
633
+ "step": 1780
634
+ },
635
+ {
636
+ "epoch": 0.12133468149646107,
637
+ "grad_norm": 0.3515625,
638
+ "learning_rate": 0.0007,
639
+ "loss": 2.286345672607422,
640
+ "step": 1800
641
+ },
642
+ {
643
+ "epoch": 0.12268284462419952,
644
+ "grad_norm": 0.396484375,
645
+ "learning_rate": 0.0007,
646
+ "loss": 2.278194236755371,
647
+ "step": 1820
648
+ },
649
+ {
650
+ "epoch": 0.12403100775193798,
651
+ "grad_norm": 0.345703125,
652
+ "learning_rate": 0.0007,
653
+ "loss": 2.245047760009766,
654
+ "step": 1840
655
+ },
656
+ {
657
+ "epoch": 0.12537917087967643,
658
+ "grad_norm": 0.369140625,
659
+ "learning_rate": 0.0007,
660
+ "loss": 2.2449272155761717,
661
+ "step": 1860
662
+ },
663
+ {
664
+ "epoch": 0.1267273340074149,
665
+ "grad_norm": 0.32421875,
666
+ "learning_rate": 0.0007,
667
+ "loss": 2.236011505126953,
668
+ "step": 1880
669
+ },
670
+ {
671
+ "epoch": 0.12807549713515334,
672
+ "grad_norm": 0.353515625,
673
+ "learning_rate": 0.0007,
674
+ "loss": 2.2427194595336912,
675
+ "step": 1900
676
+ },
677
+ {
678
+ "epoch": 0.12942366026289182,
679
+ "grad_norm": 0.341796875,
680
+ "learning_rate": 0.0007,
681
+ "loss": 2.220064163208008,
682
+ "step": 1920
683
+ },
684
+ {
685
+ "epoch": 0.13077182339063026,
686
+ "grad_norm": 0.349609375,
687
+ "learning_rate": 0.0007,
688
+ "loss": 2.230933952331543,
689
+ "step": 1940
690
+ },
691
+ {
692
+ "epoch": 0.13211998651836873,
693
+ "grad_norm": 0.345703125,
694
+ "learning_rate": 0.0007,
695
+ "loss": 2.2043487548828127,
696
+ "step": 1960
697
+ },
698
+ {
699
+ "epoch": 0.13346814964610718,
700
+ "grad_norm": 0.35546875,
701
+ "learning_rate": 0.0007,
702
+ "loss": 2.226978874206543,
703
+ "step": 1980
704
+ },
705
+ {
706
+ "epoch": 0.13481631277384565,
707
+ "grad_norm": 0.341796875,
708
+ "learning_rate": 0.0007,
709
+ "loss": 2.201649284362793,
710
+ "step": 2000
711
+ },
712
+ {
713
+ "epoch": 0.1361644759015841,
714
+ "grad_norm": 0.3515625,
715
+ "learning_rate": 0.0007,
716
+ "loss": 2.2095453262329103,
717
+ "step": 2020
718
+ },
719
+ {
720
+ "epoch": 0.13751263902932254,
721
+ "grad_norm": 0.376953125,
722
+ "learning_rate": 0.0007,
723
+ "loss": 2.189457893371582,
724
+ "step": 2040
725
+ },
726
+ {
727
+ "epoch": 0.138860802157061,
728
+ "grad_norm": 0.349609375,
729
+ "learning_rate": 0.0007,
730
+ "loss": 2.2083295822143554,
731
+ "step": 2060
732
+ },
733
+ {
734
+ "epoch": 0.14020896528479945,
735
+ "grad_norm": 0.328125,
736
+ "learning_rate": 0.0007,
737
+ "loss": 2.1837480545043944,
738
+ "step": 2080
739
+ },
740
+ {
741
+ "epoch": 0.14155712841253792,
742
+ "grad_norm": 0.330078125,
743
+ "learning_rate": 0.0007,
744
+ "loss": 2.1739730834960938,
745
+ "step": 2100
746
+ },
747
+ {
748
+ "epoch": 0.14290529154027637,
749
+ "grad_norm": 0.34375,
750
+ "learning_rate": 0.0007,
751
+ "loss": 2.1836475372314452,
752
+ "step": 2120
753
+ },
754
+ {
755
+ "epoch": 0.14425345466801484,
756
+ "grad_norm": 0.33984375,
757
+ "learning_rate": 0.0007,
758
+ "loss": 2.185003089904785,
759
+ "step": 2140
760
+ },
761
+ {
762
+ "epoch": 0.14560161779575329,
763
+ "grad_norm": 0.333984375,
764
+ "learning_rate": 0.0007,
765
+ "loss": 2.1692100524902345,
766
+ "step": 2160
767
+ },
768
+ {
769
+ "epoch": 0.14694978092349173,
770
+ "grad_norm": 0.333984375,
771
+ "learning_rate": 0.0007,
772
+ "loss": 2.17752685546875,
773
+ "step": 2180
774
+ },
775
+ {
776
+ "epoch": 0.1482979440512302,
777
+ "grad_norm": 0.32421875,
778
+ "learning_rate": 0.0007,
779
+ "loss": 2.152605438232422,
780
+ "step": 2200
781
+ }
782
+ ],
783
+ "logging_steps": 20,
784
+ "max_steps": 2500,
785
+ "num_input_tokens_seen": 0,
786
+ "num_train_epochs": 1,
787
+ "save_steps": 100,
788
+ "stateful_callbacks": {
789
+ "TrainerControl": {
790
+ "args": {
791
+ "should_epoch_stop": false,
792
+ "should_evaluate": false,
793
+ "should_log": false,
794
+ "should_save": true,
795
+ "should_training_stop": false
796
+ },
797
+ "attributes": {}
798
+ }
799
+ },
800
+ "total_flos": 7097824365772800.0,
801
+ "train_batch_size": 64,
802
+ "trial_name": null,
803
+ "trial_params": null
804
+ }
zain/Activation/out/glu-situglu-100L_run/checkpoint-2200/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:26b99898fe48ff0e75b1f0139845f22400023db32e87c8fca3404fda1b772936
3
+ size 4920
zain/Activation/out/glu-situglu-100L_run/checkpoint-2300/config.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation": "situglu",
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": 100,
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-situglu-100L_run/checkpoint-2300/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5867a1242bf3243cd78e2a9b922fb4c1d9d26badb3accbf78c4d2eab6d6993a7
3
+ size 33967272
zain/Activation/out/glu-situglu-100L_run/checkpoint-2300/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e8a47373827ddc87ca1348a26a873a948f6f0507e1f71d946b1452dd192a4c3c
3
+ size 68504996
zain/Activation/out/glu-situglu-100L_run/checkpoint-2300/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-situglu-100L_run/checkpoint-2300/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9b9bcfc4abc1cd7f19c4260bd52451981a53bf8db4a664a25c46ab7065952dd0
3
+ size 1064