Hugging Face's logo

w-ahmad
/
STRESS-TEST2

Safetensors
Model card Files
xet
 Community
w-ahmad commited on
Commit
bf377e2
·
verified ·
1 Parent(s): 79a17ab

Auto upload zain 2026-08-14T23:07:46.851605 (part 3)

Browse files
Files changed (42) hide show
  1. zain/Activation/out/glu-waleed-21L_run/checkpoint-600/config.json +36 -0
  2. zain/Activation/out/glu-waleed-21L_run/checkpoint-600/model.safetensors +3 -0
  3. zain/Activation/out/glu-waleed-21L_run/checkpoint-600/optimizer.pt +3 -0
  4. zain/Activation/out/glu-waleed-21L_run/checkpoint-600/rng_state.pth +3 -0
  5. zain/Activation/out/glu-waleed-21L_run/checkpoint-600/scheduler.pt +3 -0
  6. zain/Activation/out/glu-waleed-21L_run/checkpoint-600/tokenizer.json +0 -0
  7. zain/Activation/out/glu-waleed-21L_run/checkpoint-600/tokenizer_config.json +13 -0
  8. zain/Activation/out/glu-waleed-21L_run/checkpoint-600/trainer_state.json +244 -0
  9. zain/Activation/out/glu-waleed-21L_run/checkpoint-600/training_args.bin +3 -0
  10. zain/Activation/out/glu-waleed-21L_run/checkpoint-700/config.json +36 -0
  11. zain/Activation/out/glu-waleed-21L_run/checkpoint-700/model.safetensors +3 -0
  12. zain/Activation/out/glu-waleed-21L_run/checkpoint-700/optimizer.pt +3 -0
  13. zain/Activation/out/glu-waleed-21L_run/checkpoint-700/rng_state.pth +3 -0
  14. zain/Activation/out/glu-waleed-21L_run/checkpoint-700/scheduler.pt +3 -0
  15. zain/Activation/out/glu-waleed-21L_run/checkpoint-700/tokenizer.json +0 -0
  16. zain/Activation/out/glu-waleed-21L_run/checkpoint-700/tokenizer_config.json +13 -0
  17. zain/Activation/out/glu-waleed-21L_run/checkpoint-700/trainer_state.json +279 -0
  18. zain/Activation/out/glu-waleed-21L_run/checkpoint-700/training_args.bin +3 -0
  19. zain/Activation/out/glu-waleed-21L_run/checkpoint-800/config.json +36 -0
  20. zain/Activation/out/glu-waleed-21L_run/checkpoint-800/model.safetensors +3 -0
  21. zain/Activation/out/glu-waleed-21L_run/checkpoint-800/optimizer.pt +3 -0
  22. zain/Activation/out/glu-waleed-21L_run/checkpoint-800/rng_state.pth +3 -0
  23. zain/Activation/out/glu-waleed-21L_run/checkpoint-800/scheduler.pt +3 -0
  24. zain/Activation/out/glu-waleed-21L_run/checkpoint-800/tokenizer.json +0 -0
  25. zain/Activation/out/glu-waleed-21L_run/checkpoint-800/tokenizer_config.json +13 -0
  26. zain/Activation/out/glu-waleed-21L_run/checkpoint-800/trainer_state.json +314 -0
  27. zain/Activation/out/glu-waleed-21L_run/checkpoint-800/training_args.bin +3 -0
  28. zain/Activation/out/glu-waleed-21L_run/checkpoint-900/config.json +36 -0
  29. zain/Activation/out/glu-waleed-21L_run/checkpoint-900/model.safetensors +3 -0
  30. zain/Activation/out/glu-waleed-21L_run/checkpoint-900/optimizer.pt +3 -0
  31. zain/Activation/out/glu-waleed-21L_run/checkpoint-900/rng_state.pth +3 -0
  32. zain/Activation/out/glu-waleed-21L_run/checkpoint-900/scheduler.pt +3 -0
  33. zain/Activation/out/glu-waleed-21L_run/checkpoint-900/tokenizer.json +0 -0
  34. zain/Activation/out/glu-waleed-21L_run/checkpoint-900/tokenizer_config.json +13 -0
  35. zain/Activation/out/glu-waleed-21L_run/checkpoint-900/trainer_state.json +349 -0
  36. zain/Activation/out/glu-waleed-21L_run/checkpoint-900/training_args.bin +3 -0
  37. zain/Activation/out/glu-waleed-21L_run/config.json +36 -0
  38. zain/Activation/out/glu-waleed-21L_run/model.safetensors +3 -0
  39. zain/Activation/out/glu-waleed-21L_run/tokenizer.json +0 -0
  40. zain/Activation/out/glu-waleed-21L_run/tokenizer_config.json +13 -0
  41. zain/Activation/out/glu-waleed-21L_run/training_args.bin +3 -0
  42. zain/Activation/out/glu-waleed-21L_run/training_log.jsonl +53 -0
zain/Activation/out/glu-waleed-21L_run/checkpoint-600/config.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation": "waleed",
3
+ "architectures": [
4
+ "TinyLlamaForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 1,
9
+ "dtype": "bfloat16",
10
+ "eos_token_id": 2,
11
+ "head_dim": 32,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 128,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 256,
16
+ "max_position_embeddings": 512,
17
+ "mlp_bias": false,
18
+ "mlp_type": "glu",
19
+ "model_type": "tiny_llama",
20
+ "num_attention_heads": 4,
21
+ "num_hidden_layers": 21,
22
+ "num_key_value_heads": 4,
23
+ "pad_token_id": 0,
24
+ "pretraining_tp": 1,
25
+ "rms_norm_eps": 1e-06,
26
+ "rope_parameters": {
27
+ "rope_theta": 10000.0,
28
+ "rope_type": "default"
29
+ },
30
+ "tie_word_embeddings": true,
31
+ "tokenizer_name": "w-ahmad/tiny-stories-tokenizer",
32
+ "transformers_version": "5.16.0.dev0",
33
+ "use_cache": false,
34
+ "vocab_size": 4096,
35
+ "waleed_beta": 10.0
36
+ }
zain/Activation/out/glu-waleed-21L_run/checkpoint-600/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d76c829d9b0d971343ede7a1764befdba70a74e3e396233d10f1d60396d115e4
3
+ size 7961528
zain/Activation/out/glu-waleed-21L_run/checkpoint-600/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:57ad385d1ad4a04d9ff168078a64394a8d60b53d9d712bf050f4cd09b3959832
3
+ size 16041274
zain/Activation/out/glu-waleed-21L_run/checkpoint-600/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-waleed-21L_run/checkpoint-600/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c5502244c7ba8b61ab0a3b01cdc7e5e2629060a0565d6b782445d6ba31022cee
3
+ size 1064
zain/Activation/out/glu-waleed-21L_run/checkpoint-600/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
zain/Activation/out/glu-waleed-21L_run/checkpoint-600/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-waleed-21L_run/checkpoint-600/trainer_state.json ADDED
@@ -0,0 +1,244 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 0.3235373416015098,
6
+ "eval_steps": 60000,
7
+ "global_step": 600,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.010784578053383662,
14
+ "grad_norm": 4.0,
15
+ "learning_rate": 0.001,
16
+ "loss": 28.87608642578125,
17
+ "step": 20
18
+ },
19
+ {
20
+ "epoch": 0.021569156106767323,
21
+ "grad_norm": 1.96875,
22
+ "learning_rate": 0.001,
23
+ "loss": 24.005902099609376,
24
+ "step": 40
25
+ },
26
+ {
27
+ "epoch": 0.032353734160150985,
28
+ "grad_norm": 1.9921875,
29
+ "learning_rate": 0.001,
30
+ "loss": 22.247718811035156,
31
+ "step": 60
32
+ },
33
+ {
34
+ "epoch": 0.043138312213534646,
35
+ "grad_norm": 1.625,
36
+ "learning_rate": 0.001,
37
+ "loss": 20.391375732421874,
38
+ "step": 80
39
+ },
40
+ {
41
+ "epoch": 0.05392289026691831,
42
+ "grad_norm": 2.0625,
43
+ "learning_rate": 0.001,
44
+ "loss": 18.540757751464845,
45
+ "step": 100
46
+ },
47
+ {
48
+ "epoch": 0.06470746832030197,
49
+ "grad_norm": 2.09375,
50
+ "learning_rate": 0.001,
51
+ "loss": 17.22874755859375,
52
+ "step": 120
53
+ },
54
+ {
55
+ "epoch": 0.07549204637368563,
56
+ "grad_norm": 2.9375,
57
+ "learning_rate": 0.001,
58
+ "loss": 16.200546264648438,
59
+ "step": 140
60
+ },
61
+ {
62
+ "epoch": 0.08627662442706929,
63
+ "grad_norm": 1.3515625,
64
+ "learning_rate": 0.001,
65
+ "loss": 15.420201110839844,
66
+ "step": 160
67
+ },
68
+ {
69
+ "epoch": 0.09706120248045295,
70
+ "grad_norm": 2.296875,
71
+ "learning_rate": 0.001,
72
+ "loss": 14.777641296386719,
73
+ "step": 180
74
+ },
75
+ {
76
+ "epoch": 0.10784578053383662,
77
+ "grad_norm": 2.234375,
78
+ "learning_rate": 0.001,
79
+ "loss": 14.204789733886718,
80
+ "step": 200
81
+ },
82
+ {
83
+ "epoch": 0.11863035858722028,
84
+ "grad_norm": 2.484375,
85
+ "learning_rate": 0.001,
86
+ "loss": 13.72137908935547,
87
+ "step": 220
88
+ },
89
+ {
90
+ "epoch": 0.12941493664060394,
91
+ "grad_norm": 1.75,
92
+ "learning_rate": 0.001,
93
+ "loss": 13.307633972167968,
94
+ "step": 240
95
+ },
96
+ {
97
+ "epoch": 0.1401995146939876,
98
+ "grad_norm": 2.546875,
99
+ "learning_rate": 0.001,
100
+ "loss": 12.885848999023438,
101
+ "step": 260
102
+ },
103
+ {
104
+ "epoch": 0.15098409274737126,
105
+ "grad_norm": 2.4375,
106
+ "learning_rate": 0.001,
107
+ "loss": 12.522630310058593,
108
+ "step": 280
109
+ },
110
+ {
111
+ "epoch": 0.1617686708007549,
112
+ "grad_norm": 2.546875,
113
+ "learning_rate": 0.001,
114
+ "loss": 12.175682830810548,
115
+ "step": 300
116
+ },
117
+ {
118
+ "epoch": 0.17255324885413859,
119
+ "grad_norm": 2.09375,
120
+ "learning_rate": 0.001,
121
+ "loss": 11.860960388183594,
122
+ "step": 320
123
+ },
124
+ {
125
+ "epoch": 0.18333782690752223,
126
+ "grad_norm": 1.9921875,
127
+ "learning_rate": 0.001,
128
+ "loss": 11.57656021118164,
129
+ "step": 340
130
+ },
131
+ {
132
+ "epoch": 0.1941224049609059,
133
+ "grad_norm": 2.1875,
134
+ "learning_rate": 0.001,
135
+ "loss": 11.320143127441407,
136
+ "step": 360
137
+ },
138
+ {
139
+ "epoch": 0.20490698301428956,
140
+ "grad_norm": 2.21875,
141
+ "learning_rate": 0.001,
142
+ "loss": 11.070974731445313,
143
+ "step": 380
144
+ },
145
+ {
146
+ "epoch": 0.21569156106767323,
147
+ "grad_norm": 2.234375,
148
+ "learning_rate": 0.001,
149
+ "loss": 10.860658264160156,
150
+ "step": 400
151
+ },
152
+ {
153
+ "epoch": 0.22647613912105688,
154
+ "grad_norm": 2.515625,
155
+ "learning_rate": 0.001,
156
+ "loss": 10.65341796875,
157
+ "step": 420
158
+ },
159
+ {
160
+ "epoch": 0.23726071717444056,
161
+ "grad_norm": 2.65625,
162
+ "learning_rate": 0.001,
163
+ "loss": 10.532276916503907,
164
+ "step": 440
165
+ },
166
+ {
167
+ "epoch": 0.2480452952278242,
168
+ "grad_norm": 2.015625,
169
+ "learning_rate": 0.001,
170
+ "loss": 10.340470123291016,
171
+ "step": 460
172
+ },
173
+ {
174
+ "epoch": 0.2588298732812079,
175
+ "grad_norm": 2.28125,
176
+ "learning_rate": 0.001,
177
+ "loss": 10.183999633789062,
178
+ "step": 480
179
+ },
180
+ {
181
+ "epoch": 0.26961445133459155,
182
+ "grad_norm": 2.421875,
183
+ "learning_rate": 0.001,
184
+ "loss": 10.042955780029297,
185
+ "step": 500
186
+ },
187
+ {
188
+ "epoch": 0.2803990293879752,
189
+ "grad_norm": 1.9765625,
190
+ "learning_rate": 0.001,
191
+ "loss": 9.926879119873046,
192
+ "step": 520
193
+ },
194
+ {
195
+ "epoch": 0.29118360744135885,
196
+ "grad_norm": 2.171875,
197
+ "learning_rate": 0.001,
198
+ "loss": 9.819815063476563,
199
+ "step": 540
200
+ },
201
+ {
202
+ "epoch": 0.3019681854947425,
203
+ "grad_norm": 2.59375,
204
+ "learning_rate": 0.001,
205
+ "loss": 9.707681274414062,
206
+ "step": 560
207
+ },
208
+ {
209
+ "epoch": 0.3127527635481262,
210
+ "grad_norm": 2.109375,
211
+ "learning_rate": 0.001,
212
+ "loss": 9.596572113037109,
213
+ "step": 580
214
+ },
215
+ {
216
+ "epoch": 0.3235373416015098,
217
+ "grad_norm": 2.546875,
218
+ "learning_rate": 0.001,
219
+ "loss": 9.495830535888672,
220
+ "step": 600
221
+ }
222
+ ],
223
+ "logging_steps": 20,
224
+ "max_steps": 1000,
225
+ "num_input_tokens_seen": 0,
226
+ "num_train_epochs": 1,
227
+ "save_steps": 100,
228
+ "stateful_callbacks": {
229
+ "TrainerControl": {
230
+ "args": {
231
+ "should_epoch_stop": false,
232
+ "should_evaluate": false,
233
+ "should_log": false,
234
+ "should_save": true,
235
+ "should_training_stop": false
236
+ },
237
+ "attributes": {}
238
+ }
239
+ },
240
+ "total_flos": 3252189501849600.0,
241
+ "train_batch_size": 128,
242
+ "trial_name": null,
243
+ "trial_params": null
244
+ }
zain/Activation/out/glu-waleed-21L_run/checkpoint-600/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:184e260fa1c199e800dd4f9bd8c5f67eb018d939094a9714cd094900c0861fc4
3
+ size 4920
zain/Activation/out/glu-waleed-21L_run/checkpoint-700/config.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation": "waleed",
3
+ "architectures": [
4
+ "TinyLlamaForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 1,
9
+ "dtype": "bfloat16",
10
+ "eos_token_id": 2,
11
+ "head_dim": 32,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 128,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 256,
16
+ "max_position_embeddings": 512,
17
+ "mlp_bias": false,
18
+ "mlp_type": "glu",
19
+ "model_type": "tiny_llama",
20
+ "num_attention_heads": 4,
21
+ "num_hidden_layers": 21,
22
+ "num_key_value_heads": 4,
23
+ "pad_token_id": 0,
24
+ "pretraining_tp": 1,
25
+ "rms_norm_eps": 1e-06,
26
+ "rope_parameters": {
27
+ "rope_theta": 10000.0,
28
+ "rope_type": "default"
29
+ },
30
+ "tie_word_embeddings": true,
31
+ "tokenizer_name": "w-ahmad/tiny-stories-tokenizer",
32
+ "transformers_version": "5.16.0.dev0",
33
+ "use_cache": false,
34
+ "vocab_size": 4096,
35
+ "waleed_beta": 10.0
36
+ }
zain/Activation/out/glu-waleed-21L_run/checkpoint-700/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1d791fcbbfa93553c580febd3736b80c814e6385523deace0212c276e3ece5a6
3
+ size 7961528
zain/Activation/out/glu-waleed-21L_run/checkpoint-700/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b777c72ded966cf318a2eb31ade735bb65c649a8863c43a78909b721a6be1d31
3
+ size 16041274
zain/Activation/out/glu-waleed-21L_run/checkpoint-700/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-waleed-21L_run/checkpoint-700/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3173cd6908322b02fdacbfeb79bbb6a1a1a40d218262cceb79ee8c11c4fa07d7
3
+ size 1064
zain/Activation/out/glu-waleed-21L_run/checkpoint-700/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
zain/Activation/out/glu-waleed-21L_run/checkpoint-700/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-waleed-21L_run/checkpoint-700/trainer_state.json ADDED
@@ -0,0 +1,279 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 0.37746023186842814,
6
+ "eval_steps": 60000,
7
+ "global_step": 700,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.010784578053383662,
14
+ "grad_norm": 4.0,
15
+ "learning_rate": 0.001,
16
+ "loss": 28.87608642578125,
17
+ "step": 20
18
+ },
19
+ {
20
+ "epoch": 0.021569156106767323,
21
+ "grad_norm": 1.96875,
22
+ "learning_rate": 0.001,
23
+ "loss": 24.005902099609376,
24
+ "step": 40
25
+ },
26
+ {
27
+ "epoch": 0.032353734160150985,
28
+ "grad_norm": 1.9921875,
29
+ "learning_rate": 0.001,
30
+ "loss": 22.247718811035156,
31
+ "step": 60
32
+ },
33
+ {
34
+ "epoch": 0.043138312213534646,
35
+ "grad_norm": 1.625,
36
+ "learning_rate": 0.001,
37
+ "loss": 20.391375732421874,
38
+ "step": 80
39
+ },
40
+ {
41
+ "epoch": 0.05392289026691831,
42
+ "grad_norm": 2.0625,
43
+ "learning_rate": 0.001,
44
+ "loss": 18.540757751464845,
45
+ "step": 100
46
+ },
47
+ {
48
+ "epoch": 0.06470746832030197,
49
+ "grad_norm": 2.09375,
50
+ "learning_rate": 0.001,
51
+ "loss": 17.22874755859375,
52
+ "step": 120
53
+ },
54
+ {
55
+ "epoch": 0.07549204637368563,
56
+ "grad_norm": 2.9375,
57
+ "learning_rate": 0.001,
58
+ "loss": 16.200546264648438,
59
+ "step": 140
60
+ },
61
+ {
62
+ "epoch": 0.08627662442706929,
63
+ "grad_norm": 1.3515625,
64
+ "learning_rate": 0.001,
65
+ "loss": 15.420201110839844,
66
+ "step": 160
67
+ },
68
+ {
69
+ "epoch": 0.09706120248045295,
70
+ "grad_norm": 2.296875,
71
+ "learning_rate": 0.001,
72
+ "loss": 14.777641296386719,
73
+ "step": 180
74
+ },
75
+ {
76
+ "epoch": 0.10784578053383662,
77
+ "grad_norm": 2.234375,
78
+ "learning_rate": 0.001,
79
+ "loss": 14.204789733886718,
80
+ "step": 200
81
+ },
82
+ {
83
+ "epoch": 0.11863035858722028,
84
+ "grad_norm": 2.484375,
85
+ "learning_rate": 0.001,
86
+ "loss": 13.72137908935547,
87
+ "step": 220
88
+ },
89
+ {
90
+ "epoch": 0.12941493664060394,
91
+ "grad_norm": 1.75,
92
+ "learning_rate": 0.001,
93
+ "loss": 13.307633972167968,
94
+ "step": 240
95
+ },
96
+ {
97
+ "epoch": 0.1401995146939876,
98
+ "grad_norm": 2.546875,
99
+ "learning_rate": 0.001,
100
+ "loss": 12.885848999023438,
101
+ "step": 260
102
+ },
103
+ {
104
+ "epoch": 0.15098409274737126,
105
+ "grad_norm": 2.4375,
106
+ "learning_rate": 0.001,
107
+ "loss": 12.522630310058593,
108
+ "step": 280
109
+ },
110
+ {
111
+ "epoch": 0.1617686708007549,
112
+ "grad_norm": 2.546875,
113
+ "learning_rate": 0.001,
114
+ "loss": 12.175682830810548,
115
+ "step": 300
116
+ },
117
+ {
118
+ "epoch": 0.17255324885413859,
119
+ "grad_norm": 2.09375,
120
+ "learning_rate": 0.001,
121
+ "loss": 11.860960388183594,
122
+ "step": 320
123
+ },
124
+ {
125
+ "epoch": 0.18333782690752223,
126
+ "grad_norm": 1.9921875,
127
+ "learning_rate": 0.001,
128
+ "loss": 11.57656021118164,
129
+ "step": 340
130
+ },
131
+ {
132
+ "epoch": 0.1941224049609059,
133
+ "grad_norm": 2.1875,
134
+ "learning_rate": 0.001,
135
+ "loss": 11.320143127441407,
136
+ "step": 360
137
+ },
138
+ {
139
+ "epoch": 0.20490698301428956,
140
+ "grad_norm": 2.21875,
141
+ "learning_rate": 0.001,
142
+ "loss": 11.070974731445313,
143
+ "step": 380
144
+ },
145
+ {
146
+ "epoch": 0.21569156106767323,
147
+ "grad_norm": 2.234375,
148
+ "learning_rate": 0.001,
149
+ "loss": 10.860658264160156,
150
+ "step": 400
151
+ },
152
+ {
153
+ "epoch": 0.22647613912105688,
154
+ "grad_norm": 2.515625,
155
+ "learning_rate": 0.001,
156
+ "loss": 10.65341796875,
157
+ "step": 420
158
+ },
159
+ {
160
+ "epoch": 0.23726071717444056,
161
+ "grad_norm": 2.65625,
162
+ "learning_rate": 0.001,
163
+ "loss": 10.532276916503907,
164
+ "step": 440
165
+ },
166
+ {
167
+ "epoch": 0.2480452952278242,
168
+ "grad_norm": 2.015625,
169
+ "learning_rate": 0.001,
170
+ "loss": 10.340470123291016,
171
+ "step": 460
172
+ },
173
+ {
174
+ "epoch": 0.2588298732812079,
175
+ "grad_norm": 2.28125,
176
+ "learning_rate": 0.001,
177
+ "loss": 10.183999633789062,
178
+ "step": 480
179
+ },
180
+ {
181
+ "epoch": 0.26961445133459155,
182
+ "grad_norm": 2.421875,
183
+ "learning_rate": 0.001,
184
+ "loss": 10.042955780029297,
185
+ "step": 500
186
+ },
187
+ {
188
+ "epoch": 0.2803990293879752,
189
+ "grad_norm": 1.9765625,
190
+ "learning_rate": 0.001,
191
+ "loss": 9.926879119873046,
192
+ "step": 520
193
+ },
194
+ {
195
+ "epoch": 0.29118360744135885,
196
+ "grad_norm": 2.171875,
197
+ "learning_rate": 0.001,
198
+ "loss": 9.819815063476563,
199
+ "step": 540
200
+ },
201
+ {
202
+ "epoch": 0.3019681854947425,
203
+ "grad_norm": 2.59375,
204
+ "learning_rate": 0.001,
205
+ "loss": 9.707681274414062,
206
+ "step": 560
207
+ },
208
+ {
209
+ "epoch": 0.3127527635481262,
210
+ "grad_norm": 2.109375,
211
+ "learning_rate": 0.001,
212
+ "loss": 9.596572113037109,
213
+ "step": 580
214
+ },
215
+ {
216
+ "epoch": 0.3235373416015098,
217
+ "grad_norm": 2.546875,
218
+ "learning_rate": 0.001,
219
+ "loss": 9.495830535888672,
220
+ "step": 600
221
+ },
222
+ {
223
+ "epoch": 0.3343219196548935,
224
+ "grad_norm": 2.21875,
225
+ "learning_rate": 0.001,
226
+ "loss": 9.410254669189452,
227
+ "step": 620
228
+ },
229
+ {
230
+ "epoch": 0.34510649770827717,
231
+ "grad_norm": 2.3125,
232
+ "learning_rate": 0.001,
233
+ "loss": 9.315697479248048,
234
+ "step": 640
235
+ },
236
+ {
237
+ "epoch": 0.35589107576166085,
238
+ "grad_norm": 1.9453125,
239
+ "learning_rate": 0.001,
240
+ "loss": 9.232810974121094,
241
+ "step": 660
242
+ },
243
+ {
244
+ "epoch": 0.36667565381504447,
245
+ "grad_norm": 2.125,
246
+ "learning_rate": 0.001,
247
+ "loss": 9.158429718017578,
248
+ "step": 680
249
+ },
250
+ {
251
+ "epoch": 0.37746023186842814,
252
+ "grad_norm": 2.09375,
253
+ "learning_rate": 0.001,
254
+ "loss": 9.085089874267577,
255
+ "step": 700
256
+ }
257
+ ],
258
+ "logging_steps": 20,
259
+ "max_steps": 1000,
260
+ "num_input_tokens_seen": 0,
261
+ "num_train_epochs": 1,
262
+ "save_steps": 100,
263
+ "stateful_callbacks": {
264
+ "TrainerControl": {
265
+ "args": {
266
+ "should_epoch_stop": false,
267
+ "should_evaluate": false,
268
+ "should_log": false,
269
+ "should_save": true,
270
+ "should_training_stop": false
271
+ },
272
+ "attributes": {}
273
+ }
274
+ },
275
+ "total_flos": 3794221085491200.0,
276
+ "train_batch_size": 128,
277
+ "trial_name": null,
278
+ "trial_params": null
279
+ }
zain/Activation/out/glu-waleed-21L_run/checkpoint-700/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:184e260fa1c199e800dd4f9bd8c5f67eb018d939094a9714cd094900c0861fc4
3
+ size 4920
zain/Activation/out/glu-waleed-21L_run/checkpoint-800/config.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation": "waleed",
3
+ "architectures": [
4
+ "TinyLlamaForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 1,
9
+ "dtype": "bfloat16",
10
+ "eos_token_id": 2,
11
+ "head_dim": 32,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 128,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 256,
16
+ "max_position_embeddings": 512,
17
+ "mlp_bias": false,
18
+ "mlp_type": "glu",
19
+ "model_type": "tiny_llama",
20
+ "num_attention_heads": 4,
21
+ "num_hidden_layers": 21,
22
+ "num_key_value_heads": 4,
23
+ "pad_token_id": 0,
24
+ "pretraining_tp": 1,
25
+ "rms_norm_eps": 1e-06,
26
+ "rope_parameters": {
27
+ "rope_theta": 10000.0,
28
+ "rope_type": "default"
29
+ },
30
+ "tie_word_embeddings": true,
31
+ "tokenizer_name": "w-ahmad/tiny-stories-tokenizer",
32
+ "transformers_version": "5.16.0.dev0",
33
+ "use_cache": false,
34
+ "vocab_size": 4096,
35
+ "waleed_beta": 10.0
36
+ }
zain/Activation/out/glu-waleed-21L_run/checkpoint-800/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:319b587c9e9e8260c494403cb605ced90610df58fa21218f35691894f0e8ea81
3
+ size 7961528
zain/Activation/out/glu-waleed-21L_run/checkpoint-800/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:13bddc3a108a443dba2b2cb8061a3762054f7eac70512c4f3d6a5c1fe1f31d8f
3
+ size 16041274
zain/Activation/out/glu-waleed-21L_run/checkpoint-800/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-waleed-21L_run/checkpoint-800/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7334026bbd250befa466dcbe6bec8bfe84c03024263e4dea55e40698f2caa92e
3
+ size 1064
zain/Activation/out/glu-waleed-21L_run/checkpoint-800/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
zain/Activation/out/glu-waleed-21L_run/checkpoint-800/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-waleed-21L_run/checkpoint-800/trainer_state.json ADDED
@@ -0,0 +1,314 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 0.43138312213534646,
6
+ "eval_steps": 60000,
7
+ "global_step": 800,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.010784578053383662,
14
+ "grad_norm": 4.0,
15
+ "learning_rate": 0.001,
16
+ "loss": 28.87608642578125,
17
+ "step": 20
18
+ },
19
+ {
20
+ "epoch": 0.021569156106767323,
21
+ "grad_norm": 1.96875,
22
+ "learning_rate": 0.001,
23
+ "loss": 24.005902099609376,
24
+ "step": 40
25
+ },
26
+ {
27
+ "epoch": 0.032353734160150985,
28
+ "grad_norm": 1.9921875,
29
+ "learning_rate": 0.001,
30
+ "loss": 22.247718811035156,
31
+ "step": 60
32
+ },
33
+ {
34
+ "epoch": 0.043138312213534646,
35
+ "grad_norm": 1.625,
36
+ "learning_rate": 0.001,
37
+ "loss": 20.391375732421874,
38
+ "step": 80
39
+ },
40
+ {
41
+ "epoch": 0.05392289026691831,
42
+ "grad_norm": 2.0625,
43
+ "learning_rate": 0.001,
44
+ "loss": 18.540757751464845,
45
+ "step": 100
46
+ },
47
+ {
48
+ "epoch": 0.06470746832030197,
49
+ "grad_norm": 2.09375,
50
+ "learning_rate": 0.001,
51
+ "loss": 17.22874755859375,
52
+ "step": 120
53
+ },
54
+ {
55
+ "epoch": 0.07549204637368563,
56
+ "grad_norm": 2.9375,
57
+ "learning_rate": 0.001,
58
+ "loss": 16.200546264648438,
59
+ "step": 140
60
+ },
61
+ {
62
+ "epoch": 0.08627662442706929,
63
+ "grad_norm": 1.3515625,
64
+ "learning_rate": 0.001,
65
+ "loss": 15.420201110839844,
66
+ "step": 160
67
+ },
68
+ {
69
+ "epoch": 0.09706120248045295,
70
+ "grad_norm": 2.296875,
71
+ "learning_rate": 0.001,
72
+ "loss": 14.777641296386719,
73
+ "step": 180
74
+ },
75
+ {
76
+ "epoch": 0.10784578053383662,
77
+ "grad_norm": 2.234375,
78
+ "learning_rate": 0.001,
79
+ "loss": 14.204789733886718,
80
+ "step": 200
81
+ },
82
+ {
83
+ "epoch": 0.11863035858722028,
84
+ "grad_norm": 2.484375,
85
+ "learning_rate": 0.001,
86
+ "loss": 13.72137908935547,
87
+ "step": 220
88
+ },
89
+ {
90
+ "epoch": 0.12941493664060394,
91
+ "grad_norm": 1.75,
92
+ "learning_rate": 0.001,
93
+ "loss": 13.307633972167968,
94
+ "step": 240
95
+ },
96
+ {
97
+ "epoch": 0.1401995146939876,
98
+ "grad_norm": 2.546875,
99
+ "learning_rate": 0.001,
100
+ "loss": 12.885848999023438,
101
+ "step": 260
102
+ },
103
+ {
104
+ "epoch": 0.15098409274737126,
105
+ "grad_norm": 2.4375,
106
+ "learning_rate": 0.001,
107
+ "loss": 12.522630310058593,
108
+ "step": 280
109
+ },
110
+ {
111
+ "epoch": 0.1617686708007549,
112
+ "grad_norm": 2.546875,
113
+ "learning_rate": 0.001,
114
+ "loss": 12.175682830810548,
115
+ "step": 300
116
+ },
117
+ {
118
+ "epoch": 0.17255324885413859,
119
+ "grad_norm": 2.09375,
120
+ "learning_rate": 0.001,
121
+ "loss": 11.860960388183594,
122
+ "step": 320
123
+ },
124
+ {
125
+ "epoch": 0.18333782690752223,
126
+ "grad_norm": 1.9921875,
127
+ "learning_rate": 0.001,
128
+ "loss": 11.57656021118164,
129
+ "step": 340
130
+ },
131
+ {
132
+ "epoch": 0.1941224049609059,
133
+ "grad_norm": 2.1875,
134
+ "learning_rate": 0.001,
135
+ "loss": 11.320143127441407,
136
+ "step": 360
137
+ },
138
+ {
139
+ "epoch": 0.20490698301428956,
140
+ "grad_norm": 2.21875,
141
+ "learning_rate": 0.001,
142
+ "loss": 11.070974731445313,
143
+ "step": 380
144
+ },
145
+ {
146
+ "epoch": 0.21569156106767323,
147
+ "grad_norm": 2.234375,
148
+ "learning_rate": 0.001,
149
+ "loss": 10.860658264160156,
150
+ "step": 400
151
+ },
152
+ {
153
+ "epoch": 0.22647613912105688,
154
+ "grad_norm": 2.515625,
155
+ "learning_rate": 0.001,
156
+ "loss": 10.65341796875,
157
+ "step": 420
158
+ },
159
+ {
160
+ "epoch": 0.23726071717444056,
161
+ "grad_norm": 2.65625,
162
+ "learning_rate": 0.001,
163
+ "loss": 10.532276916503907,
164
+ "step": 440
165
+ },
166
+ {
167
+ "epoch": 0.2480452952278242,
168
+ "grad_norm": 2.015625,
169
+ "learning_rate": 0.001,
170
+ "loss": 10.340470123291016,
171
+ "step": 460
172
+ },
173
+ {
174
+ "epoch": 0.2588298732812079,
175
+ "grad_norm": 2.28125,
176
+ "learning_rate": 0.001,
177
+ "loss": 10.183999633789062,
178
+ "step": 480
179
+ },
180
+ {
181
+ "epoch": 0.26961445133459155,
182
+ "grad_norm": 2.421875,
183
+ "learning_rate": 0.001,
184
+ "loss": 10.042955780029297,
185
+ "step": 500
186
+ },
187
+ {
188
+ "epoch": 0.2803990293879752,
189
+ "grad_norm": 1.9765625,
190
+ "learning_rate": 0.001,
191
+ "loss": 9.926879119873046,
192
+ "step": 520
193
+ },
194
+ {
195
+ "epoch": 0.29118360744135885,
196
+ "grad_norm": 2.171875,
197
+ "learning_rate": 0.001,
198
+ "loss": 9.819815063476563,
199
+ "step": 540
200
+ },
201
+ {
202
+ "epoch": 0.3019681854947425,
203
+ "grad_norm": 2.59375,
204
+ "learning_rate": 0.001,
205
+ "loss": 9.707681274414062,
206
+ "step": 560
207
+ },
208
+ {
209
+ "epoch": 0.3127527635481262,
210
+ "grad_norm": 2.109375,
211
+ "learning_rate": 0.001,
212
+ "loss": 9.596572113037109,
213
+ "step": 580
214
+ },
215
+ {
216
+ "epoch": 0.3235373416015098,
217
+ "grad_norm": 2.546875,
218
+ "learning_rate": 0.001,
219
+ "loss": 9.495830535888672,
220
+ "step": 600
221
+ },
222
+ {
223
+ "epoch": 0.3343219196548935,
224
+ "grad_norm": 2.21875,
225
+ "learning_rate": 0.001,
226
+ "loss": 9.410254669189452,
227
+ "step": 620
228
+ },
229
+ {
230
+ "epoch": 0.34510649770827717,
231
+ "grad_norm": 2.3125,
232
+ "learning_rate": 0.001,
233
+ "loss": 9.315697479248048,
234
+ "step": 640
235
+ },
236
+ {
237
+ "epoch": 0.35589107576166085,
238
+ "grad_norm": 1.9453125,
239
+ "learning_rate": 0.001,
240
+ "loss": 9.232810974121094,
241
+ "step": 660
242
+ },
243
+ {
244
+ "epoch": 0.36667565381504447,
245
+ "grad_norm": 2.125,
246
+ "learning_rate": 0.001,
247
+ "loss": 9.158429718017578,
248
+ "step": 680
249
+ },
250
+ {
251
+ "epoch": 0.37746023186842814,
252
+ "grad_norm": 2.09375,
253
+ "learning_rate": 0.001,
254
+ "loss": 9.085089874267577,
255
+ "step": 700
256
+ },
257
+ {
258
+ "epoch": 0.3882448099218118,
259
+ "grad_norm": 1.9765625,
260
+ "learning_rate": 0.001,
261
+ "loss": 8.9949951171875,
262
+ "step": 720
263
+ },
264
+ {
265
+ "epoch": 0.3990293879751955,
266
+ "grad_norm": 2.0,
267
+ "learning_rate": 0.001,
268
+ "loss": 8.962834930419922,
269
+ "step": 740
270
+ },
271
+ {
272
+ "epoch": 0.4098139660285791,
273
+ "grad_norm": 1.8671875,
274
+ "learning_rate": 0.001,
275
+ "loss": 8.898590850830079,
276
+ "step": 760
277
+ },
278
+ {
279
+ "epoch": 0.4205985440819628,
280
+ "grad_norm": 1.890625,
281
+ "learning_rate": 0.001,
282
+ "loss": 8.85772247314453,
283
+ "step": 780
284
+ },
285
+ {
286
+ "epoch": 0.43138312213534646,
287
+ "grad_norm": 2.109375,
288
+ "learning_rate": 0.001,
289
+ "loss": 8.76031723022461,
290
+ "step": 800
291
+ }
292
+ ],
293
+ "logging_steps": 20,
294
+ "max_steps": 1000,
295
+ "num_input_tokens_seen": 0,
296
+ "num_train_epochs": 1,
297
+ "save_steps": 100,
298
+ "stateful_callbacks": {
299
+ "TrainerControl": {
300
+ "args": {
301
+ "should_epoch_stop": false,
302
+ "should_evaluate": false,
303
+ "should_log": false,
304
+ "should_save": true,
305
+ "should_training_stop": false
306
+ },
307
+ "attributes": {}
308
+ }
309
+ },
310
+ "total_flos": 4336252669132800.0,
311
+ "train_batch_size": 128,
312
+ "trial_name": null,
313
+ "trial_params": null
314
+ }
zain/Activation/out/glu-waleed-21L_run/checkpoint-800/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:184e260fa1c199e800dd4f9bd8c5f67eb018d939094a9714cd094900c0861fc4
3
+ size 4920
zain/Activation/out/glu-waleed-21L_run/checkpoint-900/config.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation": "waleed",
3
+ "architectures": [
4
+ "TinyLlamaForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 1,
9
+ "dtype": "bfloat16",
10
+ "eos_token_id": 2,
11
+ "head_dim": 32,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 128,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 256,
16
+ "max_position_embeddings": 512,
17
+ "mlp_bias": false,
18
+ "mlp_type": "glu",
19
+ "model_type": "tiny_llama",
20
+ "num_attention_heads": 4,
21
+ "num_hidden_layers": 21,
22
+ "num_key_value_heads": 4,
23
+ "pad_token_id": 0,
24
+ "pretraining_tp": 1,
25
+ "rms_norm_eps": 1e-06,
26
+ "rope_parameters": {
27
+ "rope_theta": 10000.0,
28
+ "rope_type": "default"
29
+ },
30
+ "tie_word_embeddings": true,
31
+ "tokenizer_name": "w-ahmad/tiny-stories-tokenizer",
32
+ "transformers_version": "5.16.0.dev0",
33
+ "use_cache": false,
34
+ "vocab_size": 4096,
35
+ "waleed_beta": 10.0
36
+ }
zain/Activation/out/glu-waleed-21L_run/checkpoint-900/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5035eec6c13c93fc8e3664b6d6d1ea9962411a6315b3d0c1258fa9afed295167
3
+ size 7961528
zain/Activation/out/glu-waleed-21L_run/checkpoint-900/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f9747827baaf8716d01d6c76c26e6e0e34cab9bf3b8963e4a3e7e5da02efad43
3
+ size 16041274
zain/Activation/out/glu-waleed-21L_run/checkpoint-900/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-waleed-21L_run/checkpoint-900/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6d48b00fea613f19335b87fb622648d28d7f8900963d07ba43c75df2b4dc3d4e
3
+ size 1064
zain/Activation/out/glu-waleed-21L_run/checkpoint-900/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
zain/Activation/out/glu-waleed-21L_run/checkpoint-900/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-waleed-21L_run/checkpoint-900/trainer_state.json ADDED
@@ -0,0 +1,349 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 0.4853060124022648,
6
+ "eval_steps": 60000,
7
+ "global_step": 900,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.010784578053383662,
14
+ "grad_norm": 4.0,
15
+ "learning_rate": 0.001,
16
+ "loss": 28.87608642578125,
17
+ "step": 20
18
+ },
19
+ {
20
+ "epoch": 0.021569156106767323,
21
+ "grad_norm": 1.96875,
22
+ "learning_rate": 0.001,
23
+ "loss": 24.005902099609376,
24
+ "step": 40
25
+ },
26
+ {
27
+ "epoch": 0.032353734160150985,
28
+ "grad_norm": 1.9921875,
29
+ "learning_rate": 0.001,
30
+ "loss": 22.247718811035156,
31
+ "step": 60
32
+ },
33
+ {
34
+ "epoch": 0.043138312213534646,
35
+ "grad_norm": 1.625,
36
+ "learning_rate": 0.001,
37
+ "loss": 20.391375732421874,
38
+ "step": 80
39
+ },
40
+ {
41
+ "epoch": 0.05392289026691831,
42
+ "grad_norm": 2.0625,
43
+ "learning_rate": 0.001,
44
+ "loss": 18.540757751464845,
45
+ "step": 100
46
+ },
47
+ {
48
+ "epoch": 0.06470746832030197,
49
+ "grad_norm": 2.09375,
50
+ "learning_rate": 0.001,
51
+ "loss": 17.22874755859375,
52
+ "step": 120
53
+ },
54
+ {
55
+ "epoch": 0.07549204637368563,
56
+ "grad_norm": 2.9375,
57
+ "learning_rate": 0.001,
58
+ "loss": 16.200546264648438,
59
+ "step": 140
60
+ },
61
+ {
62
+ "epoch": 0.08627662442706929,
63
+ "grad_norm": 1.3515625,
64
+ "learning_rate": 0.001,
65
+ "loss": 15.420201110839844,
66
+ "step": 160
67
+ },
68
+ {
69
+ "epoch": 0.09706120248045295,
70
+ "grad_norm": 2.296875,
71
+ "learning_rate": 0.001,
72
+ "loss": 14.777641296386719,
73
+ "step": 180
74
+ },
75
+ {
76
+ "epoch": 0.10784578053383662,
77
+ "grad_norm": 2.234375,
78
+ "learning_rate": 0.001,
79
+ "loss": 14.204789733886718,
80
+ "step": 200
81
+ },
82
+ {
83
+ "epoch": 0.11863035858722028,
84
+ "grad_norm": 2.484375,
85
+ "learning_rate": 0.001,
86
+ "loss": 13.72137908935547,
87
+ "step": 220
88
+ },
89
+ {
90
+ "epoch": 0.12941493664060394,
91
+ "grad_norm": 1.75,
92
+ "learning_rate": 0.001,
93
+ "loss": 13.307633972167968,
94
+ "step": 240
95
+ },
96
+ {
97
+ "epoch": 0.1401995146939876,
98
+ "grad_norm": 2.546875,
99
+ "learning_rate": 0.001,
100
+ "loss": 12.885848999023438,
101
+ "step": 260
102
+ },
103
+ {
104
+ "epoch": 0.15098409274737126,
105
+ "grad_norm": 2.4375,
106
+ "learning_rate": 0.001,
107
+ "loss": 12.522630310058593,
108
+ "step": 280
109
+ },
110
+ {
111
+ "epoch": 0.1617686708007549,
112
+ "grad_norm": 2.546875,
113
+ "learning_rate": 0.001,
114
+ "loss": 12.175682830810548,
115
+ "step": 300
116
+ },
117
+ {
118
+ "epoch": 0.17255324885413859,
119
+ "grad_norm": 2.09375,
120
+ "learning_rate": 0.001,
121
+ "loss": 11.860960388183594,
122
+ "step": 320
123
+ },
124
+ {
125
+ "epoch": 0.18333782690752223,
126
+ "grad_norm": 1.9921875,
127
+ "learning_rate": 0.001,
128
+ "loss": 11.57656021118164,
129
+ "step": 340
130
+ },
131
+ {
132
+ "epoch": 0.1941224049609059,
133
+ "grad_norm": 2.1875,
134
+ "learning_rate": 0.001,
135
+ "loss": 11.320143127441407,
136
+ "step": 360
137
+ },
138
+ {
139
+ "epoch": 0.20490698301428956,
140
+ "grad_norm": 2.21875,
141
+ "learning_rate": 0.001,
142
+ "loss": 11.070974731445313,
143
+ "step": 380
144
+ },
145
+ {
146
+ "epoch": 0.21569156106767323,
147
+ "grad_norm": 2.234375,
148
+ "learning_rate": 0.001,
149
+ "loss": 10.860658264160156,
150
+ "step": 400
151
+ },
152
+ {
153
+ "epoch": 0.22647613912105688,
154
+ "grad_norm": 2.515625,
155
+ "learning_rate": 0.001,
156
+ "loss": 10.65341796875,
157
+ "step": 420
158
+ },
159
+ {
160
+ "epoch": 0.23726071717444056,
161
+ "grad_norm": 2.65625,
162
+ "learning_rate": 0.001,
163
+ "loss": 10.532276916503907,
164
+ "step": 440
165
+ },
166
+ {
167
+ "epoch": 0.2480452952278242,
168
+ "grad_norm": 2.015625,
169
+ "learning_rate": 0.001,
170
+ "loss": 10.340470123291016,
171
+ "step": 460
172
+ },
173
+ {
174
+ "epoch": 0.2588298732812079,
175
+ "grad_norm": 2.28125,
176
+ "learning_rate": 0.001,
177
+ "loss": 10.183999633789062,
178
+ "step": 480
179
+ },
180
+ {
181
+ "epoch": 0.26961445133459155,
182
+ "grad_norm": 2.421875,
183
+ "learning_rate": 0.001,
184
+ "loss": 10.042955780029297,
185
+ "step": 500
186
+ },
187
+ {
188
+ "epoch": 0.2803990293879752,
189
+ "grad_norm": 1.9765625,
190
+ "learning_rate": 0.001,
191
+ "loss": 9.926879119873046,
192
+ "step": 520
193
+ },
194
+ {
195
+ "epoch": 0.29118360744135885,
196
+ "grad_norm": 2.171875,
197
+ "learning_rate": 0.001,
198
+ "loss": 9.819815063476563,
199
+ "step": 540
200
+ },
201
+ {
202
+ "epoch": 0.3019681854947425,
203
+ "grad_norm": 2.59375,
204
+ "learning_rate": 0.001,
205
+ "loss": 9.707681274414062,
206
+ "step": 560
207
+ },
208
+ {
209
+ "epoch": 0.3127527635481262,
210
+ "grad_norm": 2.109375,
211
+ "learning_rate": 0.001,
212
+ "loss": 9.596572113037109,
213
+ "step": 580
214
+ },
215
+ {
216
+ "epoch": 0.3235373416015098,
217
+ "grad_norm": 2.546875,
218
+ "learning_rate": 0.001,
219
+ "loss": 9.495830535888672,
220
+ "step": 600
221
+ },
222
+ {
223
+ "epoch": 0.3343219196548935,
224
+ "grad_norm": 2.21875,
225
+ "learning_rate": 0.001,
226
+ "loss": 9.410254669189452,
227
+ "step": 620
228
+ },
229
+ {
230
+ "epoch": 0.34510649770827717,
231
+ "grad_norm": 2.3125,
232
+ "learning_rate": 0.001,
233
+ "loss": 9.315697479248048,
234
+ "step": 640
235
+ },
236
+ {
237
+ "epoch": 0.35589107576166085,
238
+ "grad_norm": 1.9453125,
239
+ "learning_rate": 0.001,
240
+ "loss": 9.232810974121094,
241
+ "step": 660
242
+ },
243
+ {
244
+ "epoch": 0.36667565381504447,
245
+ "grad_norm": 2.125,
246
+ "learning_rate": 0.001,
247
+ "loss": 9.158429718017578,
248
+ "step": 680
249
+ },
250
+ {
251
+ "epoch": 0.37746023186842814,
252
+ "grad_norm": 2.09375,
253
+ "learning_rate": 0.001,
254
+ "loss": 9.085089874267577,
255
+ "step": 700
256
+ },
257
+ {
258
+ "epoch": 0.3882448099218118,
259
+ "grad_norm": 1.9765625,
260
+ "learning_rate": 0.001,
261
+ "loss": 8.9949951171875,
262
+ "step": 720
263
+ },
264
+ {
265
+ "epoch": 0.3990293879751955,
266
+ "grad_norm": 2.0,
267
+ "learning_rate": 0.001,
268
+ "loss": 8.962834930419922,
269
+ "step": 740
270
+ },
271
+ {
272
+ "epoch": 0.4098139660285791,
273
+ "grad_norm": 1.8671875,
274
+ "learning_rate": 0.001,
275
+ "loss": 8.898590850830079,
276
+ "step": 760
277
+ },
278
+ {
279
+ "epoch": 0.4205985440819628,
280
+ "grad_norm": 1.890625,
281
+ "learning_rate": 0.001,
282
+ "loss": 8.85772247314453,
283
+ "step": 780
284
+ },
285
+ {
286
+ "epoch": 0.43138312213534646,
287
+ "grad_norm": 2.109375,
288
+ "learning_rate": 0.001,
289
+ "loss": 8.76031723022461,
290
+ "step": 800
291
+ },
292
+ {
293
+ "epoch": 0.44216770018873014,
294
+ "grad_norm": 2.59375,
295
+ "learning_rate": 0.001,
296
+ "loss": 8.73520278930664,
297
+ "step": 820
298
+ },
299
+ {
300
+ "epoch": 0.45295227824211376,
301
+ "grad_norm": 2.25,
302
+ "learning_rate": 0.001,
303
+ "loss": 8.694710540771485,
304
+ "step": 840
305
+ },
306
+ {
307
+ "epoch": 0.46373685629549743,
308
+ "grad_norm": 1.9921875,
309
+ "learning_rate": 0.001,
310
+ "loss": 8.625717926025391,
311
+ "step": 860
312
+ },
313
+ {
314
+ "epoch": 0.4745214343488811,
315
+ "grad_norm": 1.921875,
316
+ "learning_rate": 0.001,
317
+ "loss": 8.585829162597657,
318
+ "step": 880
319
+ },
320
+ {
321
+ "epoch": 0.4853060124022648,
322
+ "grad_norm": 2.0625,
323
+ "learning_rate": 0.001,
324
+ "loss": 8.538186645507812,
325
+ "step": 900
326
+ }
327
+ ],
328
+ "logging_steps": 20,
329
+ "max_steps": 1000,
330
+ "num_input_tokens_seen": 0,
331
+ "num_train_epochs": 1,
332
+ "save_steps": 100,
333
+ "stateful_callbacks": {
334
+ "TrainerControl": {
335
+ "args": {
336
+ "should_epoch_stop": false,
337
+ "should_evaluate": false,
338
+ "should_log": false,
339
+ "should_save": true,
340
+ "should_training_stop": false
341
+ },
342
+ "attributes": {}
343
+ }
344
+ },
345
+ "total_flos": 4878284252774400.0,
346
+ "train_batch_size": 128,
347
+ "trial_name": null,
348
+ "trial_params": null
349
+ }
zain/Activation/out/glu-waleed-21L_run/checkpoint-900/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:184e260fa1c199e800dd4f9bd8c5f67eb018d939094a9714cd094900c0861fc4
3
+ size 4920
zain/Activation/out/glu-waleed-21L_run/config.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation": "waleed",
3
+ "architectures": [
4
+ "TinyLlamaForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 1,
9
+ "dtype": "bfloat16",
10
+ "eos_token_id": 2,
11
+ "head_dim": 32,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 128,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 256,
16
+ "max_position_embeddings": 512,
17
+ "mlp_bias": false,
18
+ "mlp_type": "glu",
19
+ "model_type": "tiny_llama",
20
+ "num_attention_heads": 4,
21
+ "num_hidden_layers": 21,
22
+ "num_key_value_heads": 4,
23
+ "pad_token_id": 0,
24
+ "pretraining_tp": 1,
25
+ "rms_norm_eps": 1e-06,
26
+ "rope_parameters": {
27
+ "rope_theta": 10000.0,
28
+ "rope_type": "default"
29
+ },
30
+ "tie_word_embeddings": true,
31
+ "tokenizer_name": "w-ahmad/tiny-stories-tokenizer",
32
+ "transformers_version": "5.16.0.dev0",
33
+ "use_cache": false,
34
+ "vocab_size": 4096,
35
+ "waleed_beta": 10.0
36
+ }
zain/Activation/out/glu-waleed-21L_run/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9c2f263b7cdd217534c820054f062bf53806e90bc948d86c1f1c72ac89a01aba
3
+ size 7961528
zain/Activation/out/glu-waleed-21L_run/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
zain/Activation/out/glu-waleed-21L_run/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-waleed-21L_run/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:184e260fa1c199e800dd4f9bd8c5f67eb018d939094a9714cd094900c0861fc4
3
+ size 4920
zain/Activation/out/glu-waleed-21L_run/training_log.jsonl ADDED
@@ -0,0 +1,53 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"step": 20, "epoch": 0.010784578053383662, "timestamp": 1786746313.5468752, "loss": 28.87608642578125, "grad_norm": 4.0, "learning_rate": 0.001, "train/total_time_seconds": 8.163195695728064, "train/time_per_step_avg": 0.4081597847864032, "train/epoch_time_elapsed": 15.973373930901289, "train/estimated_remaining_minutes": 6.6666098181779185}
2
+ {"step": 40, "epoch": 0.021569156106767323, "timestamp": 1786746329.4508228, "loss": 24.005902099609376, "grad_norm": 1.96875, "learning_rate": 0.001, "train/total_time_seconds": 16.285561438649893, "train/time_per_step_avg": 0.40713903596624734, "train/epoch_time_elapsed": 31.877322424203157, "train/estimated_remaining_minutes": 6.5142245754599575}
3
+ {"step": 60, "epoch": 0.032353734160150985, "timestamp": 1786746345.236916, "loss": 22.247718811035156, "grad_norm": 1.9921875, "learning_rate": 0.001, "train/total_time_seconds": 24.420032560825348, "train/time_per_step_avg": 0.40700054268042246, "train/epoch_time_elapsed": 47.663415383547544, "train/estimated_remaining_minutes": 6.376341835326619}
4
+ {"step": 80, "epoch": 0.043138312213534646, "timestamp": 1786746361.0407984, "loss": 20.391375732421874, "grad_norm": 1.625, "learning_rate": 0.001, "train/total_time_seconds": 32.55129201710224, "train/time_per_step_avg": 0.40689115021377803, "train/epoch_time_elapsed": 63.46729785948992, "train/estimated_remaining_minutes": 6.238997636611263}
5
+ {"step": 100, "epoch": 0.05392289026691831, "timestamp": 1786746376.7917583, "loss": 18.540757751464845, "grad_norm": 2.0625, "learning_rate": 0.001, "train/total_time_seconds": 40.66651140898466, "train/time_per_step_avg": 0.4066651140898466, "train/epoch_time_elapsed": 79.21825712919235, "train/estimated_remaining_minutes": 6.099976711347699}
6
+ {"step": 120, "epoch": 0.06470746832030197, "timestamp": 1786746392.640495, "loss": 17.22874755859375, "grad_norm": 2.09375, "learning_rate": 0.001, "train/total_time_seconds": 48.79844278842211, "train/time_per_step_avg": 0.40635247092694043, "train/epoch_time_elapsed": 95.06699458137155, "train/estimated_remaining_minutes": 5.9642541185849245}
7
+ {"step": 140, "epoch": 0.07549204637368563, "timestamp": 1786746408.4725933, "loss": 16.200546264648438, "grad_norm": 2.9375, "learning_rate": 0.001, "train/total_time_seconds": 56.95840273797512, "train/time_per_step_avg": 0.4067284129932523, "train/epoch_time_elapsed": 110.89909297972918, "train/estimated_remaining_minutes": 5.8314555184117385}
8
+ {"step": 160, "epoch": 0.08627662442706929, "timestamp": 1786746424.2131627, "loss": 15.420201110839844, "grad_norm": 1.3515625, "learning_rate": 0.001, "train/total_time_seconds": 65.08611023053527, "train/time_per_step_avg": 0.4066607766970992, "train/epoch_time_elapsed": 126.63966212421656, "train/estimated_remaining_minutes": 5.695034645171836}
9
+ {"step": 180, "epoch": 0.09706120248045295, "timestamp": 1786746440.0208163, "loss": 14.777641296386719, "grad_norm": 2.296875, "learning_rate": 0.001, "train/total_time_seconds": 73.20601180195808, "train/time_per_step_avg": 0.4065471978485584, "train/epoch_time_elapsed": 142.44731555134058, "train/estimated_remaining_minutes": 5.558234229407929}
10
+ {"step": 200, "epoch": 0.10784578053383662, "timestamp": 1786746455.781083, "loss": 14.204789733886718, "grad_norm": 2.234375, "learning_rate": 0.001, "train/total_time_seconds": 81.32999276742339, "train/time_per_step_avg": 0.4066348135843873, "train/epoch_time_elapsed": 158.20758191868663, "train/estimated_remaining_minutes": 5.421999517828226}
11
+ {"step": 220, "epoch": 0.11863035858722028, "timestamp": 1786746471.6626232, "loss": 13.72137908935547, "grad_norm": 2.484375, "learning_rate": 0.001, "train/total_time_seconds": 89.45487509295344, "train/time_per_step_avg": 0.40656432304531337, "train/epoch_time_elapsed": 174.08912231773138, "train/estimated_remaining_minutes": 5.28596989185634}
12
+ {"step": 240, "epoch": 0.12941493664060394, "timestamp": 1786746487.4276621, "loss": 13.307633972167968, "grad_norm": 1.75, "learning_rate": 0.001, "train/total_time_seconds": 97.59082996472716, "train/time_per_step_avg": 0.40632427226752044, "train/epoch_time_elapsed": 189.85416129603982, "train/estimated_remaining_minutes": 5.150627137027267}
13
+ {"step": 260, "epoch": 0.1401995146939876, "timestamp": 1786746503.352753, "loss": 12.885848999023438, "grad_norm": 2.546875, "learning_rate": 0.001, "train/total_time_seconds": 105.72575533390045, "train/time_per_step_avg": 0.40639645103365185, "train/epoch_time_elapsed": 205.7792522162199, "train/estimated_remaining_minutes": 5.015196086351688}
14
+ {"step": 280, "epoch": 0.15098409274737126, "timestamp": 1786746519.1782413, "loss": 12.522630310058593, "grad_norm": 2.4375, "learning_rate": 0.001, "train/total_time_seconds": 113.84835535660386, "train/time_per_step_avg": 0.40642343554645777, "train/epoch_time_elapsed": 221.60474045947194, "train/estimated_remaining_minutes": 4.879215229568738}
15
+ {"step": 300, "epoch": 0.1617686708007549, "timestamp": 1786746534.963451, "loss": 12.175682830810548, "grad_norm": 2.546875, "learning_rate": 0.001, "train/total_time_seconds": 121.94647969305515, "train/time_per_step_avg": 0.4061648692563176, "train/epoch_time_elapsed": 237.38994964584708, "train/estimated_remaining_minutes": 4.7423630991743675}
16
+ {"step": 320, "epoch": 0.17255324885413859, "timestamp": 1786746550.7116091, "loss": 11.860960388183594, "grad_norm": 2.09375, "learning_rate": 0.001, "train/total_time_seconds": 130.0585708245635, "train/time_per_step_avg": 0.4060369573161006, "train/epoch_time_elapsed": 253.13810624927282, "train/estimated_remaining_minutes": 4.606241050036624}
17
+ {"step": 340, "epoch": 0.18333782690752223, "timestamp": 1786746566.4321516, "loss": 11.57656021118164, "grad_norm": 1.9921875, "learning_rate": 0.001, "train/total_time_seconds": 138.13982947543263, "train/time_per_step_avg": 0.4054899951070547, "train/epoch_time_elapsed": 268.85864955559373, "train/estimated_remaining_minutes": 4.46922977714635}
18
+ {"step": 360, "epoch": 0.1941224049609059, "timestamp": 1786746582.1528332, "loss": 11.320143127441407, "grad_norm": 2.1875, "learning_rate": 0.001, "train/total_time_seconds": 146.23417679592967, "train/time_per_step_avg": 0.4050842146202922, "train/epoch_time_elapsed": 284.5793313011527, "train/estimated_remaining_minutes": 4.332864497657175}
19
+ {"step": 380, "epoch": 0.20490698301428956, "timestamp": 1786746597.9432573, "loss": 11.070974731445313, "grad_norm": 2.21875, "learning_rate": 0.001, "train/total_time_seconds": 154.3597538471222, "train/time_per_step_avg": 0.4051139849051833, "train/epoch_time_elapsed": 300.36975629627705, "train/estimated_remaining_minutes": 4.197502078298937}
20
+ {"step": 400, "epoch": 0.21569156106767323, "timestamp": 1786746613.8161557, "loss": 10.860658264160156, "grad_norm": 2.234375, "learning_rate": 0.001, "train/total_time_seconds": 162.48071990907192, "train/time_per_step_avg": 0.4053424021601677, "train/epoch_time_elapsed": 316.24265448749065, "train/estimated_remaining_minutes": 4.062017997726798}
21
+ {"step": 420, "epoch": 0.22647613912105688, "timestamp": 1786746629.7069266, "loss": 10.65341796875, "grad_norm": 2.515625, "learning_rate": 0.001, "train/total_time_seconds": 170.5781047642231, "train/time_per_step_avg": 0.40519533939659597, "train/epoch_time_elapsed": 332.1334254331887, "train/estimated_remaining_minutes": 3.926003998541643}
22
+ {"step": 440, "epoch": 0.23726071717444056, "timestamp": 1786746645.4986134, "loss": 10.532276916503907, "grad_norm": 2.65625, "learning_rate": 0.001, "train/total_time_seconds": 178.668152179569, "train/time_per_step_avg": 0.4052832270413637, "train/epoch_time_elapsed": 347.92511208727956, "train/estimated_remaining_minutes": 3.7899305007787367}
23
+ {"step": 460, "epoch": 0.2480452952278242, "timestamp": 1786746661.3846898, "loss": 10.340470123291016, "grad_norm": 2.015625, "learning_rate": 0.001, "train/total_time_seconds": 186.78909476473927, "train/time_per_step_avg": 0.40554917968809606, "train/epoch_time_elapsed": 363.81118861585855, "train/estimated_remaining_minutes": 3.654569245397073}
24
+ {"step": 480, "epoch": 0.2588298732812079, "timestamp": 1786746677.2647724, "loss": 10.183999633789062, "grad_norm": 2.28125, "learning_rate": 0.001, "train/total_time_seconds": 194.9149330072105, "train/time_per_step_avg": 0.405551791600883, "train/epoch_time_elapsed": 379.6912710517645, "train/estimated_remaining_minutes": 3.5192974015190783}
25
+ {"step": 500, "epoch": 0.26961445133459155, "timestamp": 1786746692.984028, "loss": 10.042955780029297, "grad_norm": 2.421875, "learning_rate": 0.001, "train/total_time_seconds": 203.05824576690793, "train/time_per_step_avg": 0.40577525857836005, "train/epoch_time_elapsed": 395.4105266518891, "train/estimated_remaining_minutes": 3.384304096115132}
26
+ {"step": 520, "epoch": 0.2803990293879752, "timestamp": 1786746708.8926647, "loss": 9.926879119873046, "grad_norm": 1.9765625, "learning_rate": 0.001, "train/total_time_seconds": 211.1958619169891, "train/time_per_step_avg": 0.4061775715276599, "train/epoch_time_elapsed": 411.31916319206357, "train/estimated_remaining_minutes": 3.2491671064152166}
27
+ {"step": 540, "epoch": 0.29118360744135885, "timestamp": 1786746724.6475203, "loss": 9.819815063476563, "grad_norm": 2.171875, "learning_rate": 0.001, "train/total_time_seconds": 219.32574209570885, "train/time_per_step_avg": 0.4065758991613984, "train/epoch_time_elapsed": 427.0740192718804, "train/estimated_remaining_minutes": 3.1138839927168545}
28
+ {"step": 560, "epoch": 0.3019681854947425, "timestamp": 1786746740.3727982, "loss": 9.707681274414062, "grad_norm": 2.59375, "learning_rate": 0.001, "train/total_time_seconds": 227.45194453001022, "train/time_per_step_avg": 0.4066284976527095, "train/epoch_time_elapsed": 442.7992974705994, "train/estimated_remaining_minutes": 2.9785373688453722}
29
+ {"step": 580, "epoch": 0.3127527635481262, "timestamp": 1786746756.1175587, "loss": 9.596572113037109, "grad_norm": 2.109375, "learning_rate": 0.001, "train/total_time_seconds": 235.54809276014566, "train/time_per_step_avg": 0.4063315975293517, "train/epoch_time_elapsed": 458.5440569706261, "train/estimated_remaining_minutes": 2.8428218091741715}
30
+ {"step": 600, "epoch": 0.3235373416015098, "timestamp": 1786746772.0347462, "loss": 9.495830535888672, "grad_norm": 2.546875, "learning_rate": 0.001, "train/total_time_seconds": 243.64490681141615, "train/time_per_step_avg": 0.4058666104450822, "train/epoch_time_elapsed": 474.46124524623156, "train/estimated_remaining_minutes": 2.7071656312379573}
31
+ {"step": 620, "epoch": 0.3343219196548935, "timestamp": 1786746787.9472218, "loss": 9.410254669189452, "grad_norm": 2.21875, "learning_rate": 0.001, "train/total_time_seconds": 251.73054047301412, "train/time_per_step_avg": 0.4053467855602503, "train/epoch_time_elapsed": 490.3737205006182, "train/estimated_remaining_minutes": 2.5714410048318643}
32
+ {"step": 640, "epoch": 0.34510649770827717, "timestamp": 1786746803.6764345, "loss": 9.315697479248048, "grad_norm": 2.3125, "learning_rate": 0.001, "train/total_time_seconds": 259.8267554566264, "train/time_per_step_avg": 0.4050101336091757, "train/epoch_time_elapsed": 506.10293316468596, "train/estimated_remaining_minutes": 2.4358758324058725}
33
+ {"step": 660, "epoch": 0.35589107576166085, "timestamp": 1786746819.6057236, "loss": 9.232810974121094, "grad_norm": 1.9453125, "learning_rate": 0.001, "train/total_time_seconds": 267.9598525315523, "train/time_per_step_avg": 0.4050790800154209, "train/epoch_time_elapsed": 522.032222867012, "train/estimated_remaining_minutes": 2.300665400523429}
34
+ {"step": 680, "epoch": 0.36667565381504447, "timestamp": 1786746835.4909947, "loss": 9.158429718017578, "grad_norm": 2.125, "learning_rate": 0.001, "train/total_time_seconds": 276.09002800285816, "train/time_per_step_avg": 0.40541935242712496, "train/epoch_time_elapsed": 537.9174937456846, "train/estimated_remaining_minutes": 2.1654119843361426}
35
+ {"step": 700, "epoch": 0.37746023186842814, "timestamp": 1786746851.3299313, "loss": 9.085089874267577, "grad_norm": 2.09375, "learning_rate": 0.001, "train/total_time_seconds": 284.22142465785146, "train/time_per_step_avg": 0.40576517846435306, "train/epoch_time_elapsed": 553.7564278915524, "train/estimated_remaining_minutes": 2.0301530332703677}
36
+ {"step": 720, "epoch": 0.3882448099218118, "timestamp": 1786746867.3291252, "loss": 8.9949951171875, "grad_norm": 1.9765625, "learning_rate": 0.001, "train/total_time_seconds": 292.3600522316992, "train/time_per_step_avg": 0.4062951175868511, "train/epoch_time_elapsed": 569.7556221187115, "train/estimated_remaining_minutes": 1.8949262644647173}
37
+ {"step": 740, "epoch": 0.3990293879751955, "timestamp": 1786746883.1942728, "loss": 8.962834930419922, "grad_norm": 2.0, "learning_rate": 0.001, "train/total_time_seconds": 300.45849879831076, "train/time_per_step_avg": 0.4063174334168434, "train/epoch_time_elapsed": 585.6207705810666, "train/estimated_remaining_minutes": 1.7594416596297477}
38
+ {"step": 760, "epoch": 0.4098139660285791, "timestamp": 1786746899.1765156, "loss": 8.898590850830079, "grad_norm": 1.8671875, "learning_rate": 0.001, "train/total_time_seconds": 308.5490870960057, "train/time_per_step_avg": 0.40589234564453364, "train/epoch_time_elapsed": 601.6030120700598, "train/estimated_remaining_minutes": 1.6239425636631877}
39
+ {"step": 780, "epoch": 0.4205985440819628, "timestamp": 1786746914.9450598, "loss": 8.85772247314453, "grad_norm": 1.890625, "learning_rate": 0.001, "train/total_time_seconds": 316.6450892724097, "train/time_per_step_avg": 0.4055506126955152, "train/epoch_time_elapsed": 617.3715577833354, "train/estimated_remaining_minutes": 1.4885025564087635}
40
+ {"step": 800, "epoch": 0.43138312213534646, "timestamp": 1786746930.8624065, "loss": 8.76031723022461, "grad_norm": 2.109375, "learning_rate": 0.001, "train/total_time_seconds": 324.7415440566838, "train/time_per_step_avg": 0.4052011939883232, "train/epoch_time_elapsed": 633.2889043055475, "train/estimated_remaining_minutes": 1.353089766902849}
41
+ {"step": 820, "epoch": 0.44216770018873014, "timestamp": 1786746946.6793406, "loss": 8.73520278930664, "grad_norm": 2.59375, "learning_rate": 0.001, "train/total_time_seconds": 332.8385955616832, "train/time_per_step_avg": 0.4047854332998395, "train/epoch_time_elapsed": 649.1058385185897, "train/estimated_remaining_minutes": 1.2177021788842066}
42
+ {"step": 840, "epoch": 0.45295227824211376, "timestamp": 1786746962.5114012, "loss": 8.694710540771485, "grad_norm": 2.25, "learning_rate": 0.001, "train/total_time_seconds": 340.9439224638045, "train/time_per_step_avg": 0.40485423665493725, "train/epoch_time_elapsed": 664.9379002004862, "train/estimated_remaining_minutes": 1.0823616586152525}
43
+ {"step": 860, "epoch": 0.46373685629549743, "timestamp": 1786746978.1563046, "loss": 8.625717926025391, "grad_norm": 1.9921875, "learning_rate": 0.001, "train/total_time_seconds": 349.054763417691, "train/time_per_step_avg": 0.40505676321685313, "train/epoch_time_elapsed": 680.5828028135002, "train/estimated_remaining_minutes": 0.9470478077224175}
44
+ {"step": 880, "epoch": 0.4745214343488811, "timestamp": 1786746994.0585952, "loss": 8.585829162597657, "grad_norm": 1.921875, "learning_rate": 0.001, "train/total_time_seconds": 357.2033238746226, "train/time_per_step_avg": 0.40558234602212906, "train/epoch_time_elapsed": 696.4850920550525, "train/estimated_remaining_minutes": 0.8118257360786877}
45
+ {"step": 900, "epoch": 0.4853060124022648, "timestamp": 1786747009.8508887, "loss": 8.538186645507812, "grad_norm": 2.0625, "learning_rate": 0.001, "train/total_time_seconds": 365.3007862865925, "train/time_per_step_avg": 0.40559242229908704, "train/epoch_time_elapsed": 712.2773864157498, "train/estimated_remaining_minutes": 0.6764829375677639}
46
+ {"step": 920, "epoch": 0.4960905904556484, "timestamp": 1786747025.7478921, "loss": 8.491840362548828, "grad_norm": 1.921875, "learning_rate": 0.001, "train/total_time_seconds": 373.4075424000621, "train/time_per_step_avg": 0.40568946838378905, "train/epoch_time_elapsed": 728.1743914373219, "train/estimated_remaining_minutes": 0.5411703513044378}
47
+ {"step": 940, "epoch": 0.5068751685090321, "timestamp": 1786747041.4513707, "loss": 8.474272918701171, "grad_norm": 1.890625, "learning_rate": 0.001, "train/total_time_seconds": 381.53393379226327, "train/time_per_step_avg": 0.40590011328458786, "train/epoch_time_elapsed": 743.8778703175485, "train/estimated_remaining_minutes": 0.40588716360879074}
48
+ {"step": 960, "epoch": 0.5176597465624158, "timestamp": 1786747057.3443298, "loss": 8.452259826660157, "grad_norm": 1.84375, "learning_rate": 0.001, "train/total_time_seconds": 389.65675657242537, "train/time_per_step_avg": 0.4060199315473437, "train/epoch_time_elapsed": 759.7708287388086, "train/estimated_remaining_minutes": 0.2705949698419621}
49
+ {"step": 980, "epoch": 0.5284443246157994, "timestamp": 1786747073.2542508, "loss": 8.417095947265626, "grad_norm": 1.890625, "learning_rate": 0.001, "train/total_time_seconds": 397.7882548831403, "train/time_per_step_avg": 0.40584931008517744, "train/epoch_time_elapsed": 775.6807492077351, "train/estimated_remaining_minutes": 0.1353021275112722}
50
+ {"step": 1000, "epoch": 0.5392289026691831, "timestamp": 1786747089.150812, "loss": 8.381436920166015, "grad_norm": 1.84375, "learning_rate": 0.001, "train/total_time_seconds": 405.9349117912352, "train/time_per_step_avg": 0.40634125504642726, "train/epoch_time_elapsed": 791.5773114152253, "train/estimated_remaining_minutes": 0.0}
51
+ {"step": 1000, "epoch": 0.5392289026691831, "timestamp": 1786747098.7421207, "eval_loss": 2.0943470001220703, "eval_runtime": 9.5898, "eval_samples_per_second": 993.447, "eval_steps_per_second": 7.821, "train/total_time_seconds": 405.9349117912352, "train/time_per_step_avg": 0.40634125504642726, "train/epoch_time_elapsed": 801.1686193794012, "train/estimated_remaining_minutes": 0.0}
52
+ {"step": 1000, "epoch": 0.5392289026691831, "timestamp": 1786747098.8224442, "train_runtime": 802.0703, "train_samples_per_second": 638.348, "train_steps_per_second": 1.247, "total_flos": 5420315836416000.0, "train_loss": 11.791388671875, "train/total_time_seconds": 405.9349117912352, "train/time_per_step_avg": 0.40634125504642726, "train/epoch_time_elapsed": 801.2489429563284, "train/estimated_remaining_minutes": 0.0}
53
+ {"step": 1000, "epoch": 0.5392289026691831, "timestamp": 1786747108.5873513, "eval_loss": 2.0943470001220703, "eval_runtime": 9.7624, "eval_samples_per_second": 975.891, "eval_steps_per_second": 7.683, "train/total_time_seconds": 405.9349117912352, "train/time_per_step_avg": 0.40634125504642726, "train/epoch_time_elapsed": 811.0138485506177, "train/estimated_remaining_minutes": 0.0}