w-ahmad commited on
Commit
b4bfbe9
·
verified ·
1 Parent(s): 32f2c91

Auto upload zain 2026-08-14T23:07:46.851605 (part 5)

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-waleed10-21L_run/checkpoint-300/training_args.bin +3 -0
  2. zain/Activation/out/glu-waleed10-21L_run/checkpoint-400/config.json +36 -0
  3. zain/Activation/out/glu-waleed10-21L_run/checkpoint-400/model.safetensors +3 -0
  4. zain/Activation/out/glu-waleed10-21L_run/checkpoint-400/optimizer.pt +3 -0
  5. zain/Activation/out/glu-waleed10-21L_run/checkpoint-400/rng_state.pth +3 -0
  6. zain/Activation/out/glu-waleed10-21L_run/checkpoint-400/scheduler.pt +3 -0
  7. zain/Activation/out/glu-waleed10-21L_run/checkpoint-400/tokenizer.json +0 -0
  8. zain/Activation/out/glu-waleed10-21L_run/checkpoint-400/tokenizer_config.json +13 -0
  9. zain/Activation/out/glu-waleed10-21L_run/checkpoint-400/trainer_state.json +174 -0
  10. zain/Activation/out/glu-waleed10-21L_run/checkpoint-400/training_args.bin +3 -0
  11. zain/Activation/out/glu-waleed10-21L_run/checkpoint-500/config.json +36 -0
  12. zain/Activation/out/glu-waleed10-21L_run/checkpoint-500/model.safetensors +3 -0
  13. zain/Activation/out/glu-waleed10-21L_run/checkpoint-500/optimizer.pt +3 -0
  14. zain/Activation/out/glu-waleed10-21L_run/checkpoint-500/rng_state.pth +3 -0
  15. zain/Activation/out/glu-waleed10-21L_run/checkpoint-500/scheduler.pt +3 -0
  16. zain/Activation/out/glu-waleed10-21L_run/checkpoint-500/tokenizer.json +0 -0
  17. zain/Activation/out/glu-waleed10-21L_run/checkpoint-500/tokenizer_config.json +13 -0
  18. zain/Activation/out/glu-waleed10-21L_run/checkpoint-500/trainer_state.json +209 -0
  19. zain/Activation/out/glu-waleed10-21L_run/checkpoint-500/training_args.bin +3 -0
  20. zain/Activation/out/glu-waleed10-21L_run/checkpoint-600/config.json +36 -0
  21. zain/Activation/out/glu-waleed10-21L_run/checkpoint-600/model.safetensors +3 -0
  22. zain/Activation/out/glu-waleed10-21L_run/checkpoint-600/optimizer.pt +3 -0
  23. zain/Activation/out/glu-waleed10-21L_run/checkpoint-600/rng_state.pth +3 -0
  24. zain/Activation/out/glu-waleed10-21L_run/checkpoint-600/scheduler.pt +3 -0
  25. zain/Activation/out/glu-waleed10-21L_run/checkpoint-600/tokenizer.json +0 -0
  26. zain/Activation/out/glu-waleed10-21L_run/checkpoint-600/tokenizer_config.json +13 -0
  27. zain/Activation/out/glu-waleed10-21L_run/checkpoint-600/trainer_state.json +244 -0
  28. zain/Activation/out/glu-waleed10-21L_run/checkpoint-600/training_args.bin +3 -0
  29. zain/Activation/out/glu-waleed10-21L_run/checkpoint-700/config.json +36 -0
  30. zain/Activation/out/glu-waleed10-21L_run/checkpoint-700/model.safetensors +3 -0
  31. zain/Activation/out/glu-waleed10-21L_run/checkpoint-700/optimizer.pt +3 -0
  32. zain/Activation/out/glu-waleed10-21L_run/checkpoint-700/rng_state.pth +3 -0
  33. zain/Activation/out/glu-waleed10-21L_run/checkpoint-700/scheduler.pt +3 -0
  34. zain/Activation/out/glu-waleed10-21L_run/checkpoint-700/tokenizer.json +0 -0
  35. zain/Activation/out/glu-waleed10-21L_run/checkpoint-700/tokenizer_config.json +13 -0
  36. zain/Activation/out/glu-waleed10-21L_run/checkpoint-700/trainer_state.json +279 -0
  37. zain/Activation/out/glu-waleed10-21L_run/checkpoint-700/training_args.bin +3 -0
  38. zain/Activation/out/glu-waleed10-21L_run/checkpoint-800/config.json +36 -0
  39. zain/Activation/out/glu-waleed10-21L_run/checkpoint-800/model.safetensors +3 -0
  40. zain/Activation/out/glu-waleed10-21L_run/checkpoint-800/optimizer.pt +3 -0
  41. zain/Activation/out/glu-waleed10-21L_run/checkpoint-800/rng_state.pth +3 -0
  42. zain/Activation/out/glu-waleed10-21L_run/checkpoint-800/scheduler.pt +3 -0
  43. zain/Activation/out/glu-waleed10-21L_run/checkpoint-800/tokenizer.json +0 -0
  44. zain/Activation/out/glu-waleed10-21L_run/checkpoint-800/tokenizer_config.json +13 -0
  45. zain/Activation/out/glu-waleed10-21L_run/checkpoint-800/trainer_state.json +314 -0
  46. zain/Activation/out/glu-waleed10-21L_run/checkpoint-800/training_args.bin +3 -0
  47. zain/Activation/out/glu-waleed10-21L_run/checkpoint-900/config.json +36 -0
  48. zain/Activation/out/glu-waleed10-21L_run/checkpoint-900/model.safetensors +3 -0
  49. zain/Activation/out/glu-waleed10-21L_run/checkpoint-900/optimizer.pt +3 -0
  50. zain/Activation/out/glu-waleed10-21L_run/checkpoint-900/rng_state.pth +3 -0
zain/Activation/out/glu-waleed10-21L_run/checkpoint-300/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d4a5fb74d92c2a824150af3ee3051d6e747acb606e30ff7dc799c234f6c87bd3
3
+ size 4920
zain/Activation/out/glu-waleed10-21L_run/checkpoint-400/config.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation": "waleed10",
3
+ "architectures": [
4
+ "TinyLlamaForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 1,
9
+ "dtype": "bfloat16",
10
+ "eos_token_id": 2,
11
+ "head_dim": 32,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 128,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 256,
16
+ "max_position_embeddings": 512,
17
+ "mlp_bias": false,
18
+ "mlp_type": "glu",
19
+ "model_type": "tiny_llama",
20
+ "num_attention_heads": 4,
21
+ "num_hidden_layers": 21,
22
+ "num_key_value_heads": 4,
23
+ "pad_token_id": 0,
24
+ "pretraining_tp": 1,
25
+ "rms_norm_eps": 1e-06,
26
+ "rope_parameters": {
27
+ "rope_theta": 10000.0,
28
+ "rope_type": "default"
29
+ },
30
+ "tie_word_embeddings": true,
31
+ "tokenizer_name": "w-ahmad/tiny-stories-tokenizer",
32
+ "transformers_version": "5.16.0.dev0",
33
+ "use_cache": false,
34
+ "vocab_size": 4096,
35
+ "waleed_beta": 10.0
36
+ }
zain/Activation/out/glu-waleed10-21L_run/checkpoint-400/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8d78057f8f3b372c8d66b1e86a22ffb668e81f74687212890aedac79eca93931
3
+ size 7961528
zain/Activation/out/glu-waleed10-21L_run/checkpoint-400/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c700f909b3bf88f6d58b3ad8b4391a6312ec383fa81e6b44bd83a9dc145943ab
3
+ size 16041274
zain/Activation/out/glu-waleed10-21L_run/checkpoint-400/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9d9cd6a0487226e5bd30d1846894c82af483733ab4381b75bae9c0745e05d405
3
+ size 14244
zain/Activation/out/glu-waleed10-21L_run/checkpoint-400/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:51769a11c87b2b665cbe64c58b934afdb1fa1998bffb4addcc2f852b172d6681
3
+ size 1064
zain/Activation/out/glu-waleed10-21L_run/checkpoint-400/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
zain/Activation/out/glu-waleed10-21L_run/checkpoint-400/tokenizer_config.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "backend": "tokenizers",
4
+ "bos_token": "<|endoftext|>",
5
+ "eos_token": "<|endoftext|>",
6
+ "errors": "replace",
7
+ "is_local": false,
8
+ "local_files_only": false,
9
+ "model_max_length": 1024,
10
+ "pad_token": "<|endoftext|>",
11
+ "tokenizer_class": "GPT2Tokenizer",
12
+ "unk_token": "<|endoftext|>"
13
+ }
zain/Activation/out/glu-waleed10-21L_run/checkpoint-400/trainer_state.json ADDED
@@ -0,0 +1,174 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 0.21569156106767323,
6
+ "eval_steps": 60000,
7
+ "global_step": 400,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.010784578053383662,
14
+ "grad_norm": 4.0,
15
+ "learning_rate": 0.001,
16
+ "loss": 28.876385498046876,
17
+ "step": 20
18
+ },
19
+ {
20
+ "epoch": 0.021569156106767323,
21
+ "grad_norm": 1.3984375,
22
+ "learning_rate": 0.001,
23
+ "loss": 24.010443115234374,
24
+ "step": 40
25
+ },
26
+ {
27
+ "epoch": 0.032353734160150985,
28
+ "grad_norm": 1.453125,
29
+ "learning_rate": 0.001,
30
+ "loss": 22.35637664794922,
31
+ "step": 60
32
+ },
33
+ {
34
+ "epoch": 0.043138312213534646,
35
+ "grad_norm": 3.34375,
36
+ "learning_rate": 0.001,
37
+ "loss": 20.596783447265626,
38
+ "step": 80
39
+ },
40
+ {
41
+ "epoch": 0.05392289026691831,
42
+ "grad_norm": 1.9140625,
43
+ "learning_rate": 0.001,
44
+ "loss": 18.65278015136719,
45
+ "step": 100
46
+ },
47
+ {
48
+ "epoch": 0.06470746832030197,
49
+ "grad_norm": 3.109375,
50
+ "learning_rate": 0.001,
51
+ "loss": 17.364892578125,
52
+ "step": 120
53
+ },
54
+ {
55
+ "epoch": 0.07549204637368563,
56
+ "grad_norm": 1.6015625,
57
+ "learning_rate": 0.001,
58
+ "loss": 16.395790100097656,
59
+ "step": 140
60
+ },
61
+ {
62
+ "epoch": 0.08627662442706929,
63
+ "grad_norm": 1.5234375,
64
+ "learning_rate": 0.001,
65
+ "loss": 15.567823791503907,
66
+ "step": 160
67
+ },
68
+ {
69
+ "epoch": 0.09706120248045295,
70
+ "grad_norm": 1.703125,
71
+ "learning_rate": 0.001,
72
+ "loss": 14.909393310546875,
73
+ "step": 180
74
+ },
75
+ {
76
+ "epoch": 0.10784578053383662,
77
+ "grad_norm": 2.4375,
78
+ "learning_rate": 0.001,
79
+ "loss": 14.338616943359375,
80
+ "step": 200
81
+ },
82
+ {
83
+ "epoch": 0.11863035858722028,
84
+ "grad_norm": 3.0625,
85
+ "learning_rate": 0.001,
86
+ "loss": 13.839776611328125,
87
+ "step": 220
88
+ },
89
+ {
90
+ "epoch": 0.12941493664060394,
91
+ "grad_norm": 1.5703125,
92
+ "learning_rate": 0.001,
93
+ "loss": 13.418243408203125,
94
+ "step": 240
95
+ },
96
+ {
97
+ "epoch": 0.1401995146939876,
98
+ "grad_norm": 1.96875,
99
+ "learning_rate": 0.001,
100
+ "loss": 12.980410766601562,
101
+ "step": 260
102
+ },
103
+ {
104
+ "epoch": 0.15098409274737126,
105
+ "grad_norm": 2.578125,
106
+ "learning_rate": 0.001,
107
+ "loss": 12.627825164794922,
108
+ "step": 280
109
+ },
110
+ {
111
+ "epoch": 0.1617686708007549,
112
+ "grad_norm": 2.078125,
113
+ "learning_rate": 0.001,
114
+ "loss": 12.301615142822266,
115
+ "step": 300
116
+ },
117
+ {
118
+ "epoch": 0.17255324885413859,
119
+ "grad_norm": 2.09375,
120
+ "learning_rate": 0.001,
121
+ "loss": 11.97824478149414,
122
+ "step": 320
123
+ },
124
+ {
125
+ "epoch": 0.18333782690752223,
126
+ "grad_norm": 2.171875,
127
+ "learning_rate": 0.001,
128
+ "loss": 11.688941955566406,
129
+ "step": 340
130
+ },
131
+ {
132
+ "epoch": 0.1941224049609059,
133
+ "grad_norm": 2.125,
134
+ "learning_rate": 0.001,
135
+ "loss": 11.42403335571289,
136
+ "step": 360
137
+ },
138
+ {
139
+ "epoch": 0.20490698301428956,
140
+ "grad_norm": 2.234375,
141
+ "learning_rate": 0.001,
142
+ "loss": 11.158559417724609,
143
+ "step": 380
144
+ },
145
+ {
146
+ "epoch": 0.21569156106767323,
147
+ "grad_norm": 2.453125,
148
+ "learning_rate": 0.001,
149
+ "loss": 10.936893463134766,
150
+ "step": 400
151
+ }
152
+ ],
153
+ "logging_steps": 20,
154
+ "max_steps": 1000,
155
+ "num_input_tokens_seen": 0,
156
+ "num_train_epochs": 1,
157
+ "save_steps": 100,
158
+ "stateful_callbacks": {
159
+ "TrainerControl": {
160
+ "args": {
161
+ "should_epoch_stop": false,
162
+ "should_evaluate": false,
163
+ "should_log": false,
164
+ "should_save": true,
165
+ "should_training_stop": false
166
+ },
167
+ "attributes": {}
168
+ }
169
+ },
170
+ "total_flos": 2168126334566400.0,
171
+ "train_batch_size": 128,
172
+ "trial_name": null,
173
+ "trial_params": null
174
+ }
zain/Activation/out/glu-waleed10-21L_run/checkpoint-400/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d4a5fb74d92c2a824150af3ee3051d6e747acb606e30ff7dc799c234f6c87bd3
3
+ size 4920
zain/Activation/out/glu-waleed10-21L_run/checkpoint-500/config.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation": "waleed10",
3
+ "architectures": [
4
+ "TinyLlamaForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 1,
9
+ "dtype": "bfloat16",
10
+ "eos_token_id": 2,
11
+ "head_dim": 32,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 128,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 256,
16
+ "max_position_embeddings": 512,
17
+ "mlp_bias": false,
18
+ "mlp_type": "glu",
19
+ "model_type": "tiny_llama",
20
+ "num_attention_heads": 4,
21
+ "num_hidden_layers": 21,
22
+ "num_key_value_heads": 4,
23
+ "pad_token_id": 0,
24
+ "pretraining_tp": 1,
25
+ "rms_norm_eps": 1e-06,
26
+ "rope_parameters": {
27
+ "rope_theta": 10000.0,
28
+ "rope_type": "default"
29
+ },
30
+ "tie_word_embeddings": true,
31
+ "tokenizer_name": "w-ahmad/tiny-stories-tokenizer",
32
+ "transformers_version": "5.16.0.dev0",
33
+ "use_cache": false,
34
+ "vocab_size": 4096,
35
+ "waleed_beta": 10.0
36
+ }
zain/Activation/out/glu-waleed10-21L_run/checkpoint-500/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:00f0ea8a46b82ffafc35df415c07a25ab95ddf8aedeeed2ab7f74ab9ca7c6eb3
3
+ size 7961528
zain/Activation/out/glu-waleed10-21L_run/checkpoint-500/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1c15c131263f0ffabbabb5221bc7cafd56fa67336ace3b0f17382f12c5408acb
3
+ size 16041274
zain/Activation/out/glu-waleed10-21L_run/checkpoint-500/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9d9cd6a0487226e5bd30d1846894c82af483733ab4381b75bae9c0745e05d405
3
+ size 14244
zain/Activation/out/glu-waleed10-21L_run/checkpoint-500/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:60a0752f1683b6bad1b285850c7a77f8c14fbcad243554c3a99f6e443a78365e
3
+ size 1064
zain/Activation/out/glu-waleed10-21L_run/checkpoint-500/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
zain/Activation/out/glu-waleed10-21L_run/checkpoint-500/tokenizer_config.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "backend": "tokenizers",
4
+ "bos_token": "<|endoftext|>",
5
+ "eos_token": "<|endoftext|>",
6
+ "errors": "replace",
7
+ "is_local": false,
8
+ "local_files_only": false,
9
+ "model_max_length": 1024,
10
+ "pad_token": "<|endoftext|>",
11
+ "tokenizer_class": "GPT2Tokenizer",
12
+ "unk_token": "<|endoftext|>"
13
+ }
zain/Activation/out/glu-waleed10-21L_run/checkpoint-500/trainer_state.json ADDED
@@ -0,0 +1,209 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 0.26961445133459155,
6
+ "eval_steps": 60000,
7
+ "global_step": 500,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.010784578053383662,
14
+ "grad_norm": 4.0,
15
+ "learning_rate": 0.001,
16
+ "loss": 28.876385498046876,
17
+ "step": 20
18
+ },
19
+ {
20
+ "epoch": 0.021569156106767323,
21
+ "grad_norm": 1.3984375,
22
+ "learning_rate": 0.001,
23
+ "loss": 24.010443115234374,
24
+ "step": 40
25
+ },
26
+ {
27
+ "epoch": 0.032353734160150985,
28
+ "grad_norm": 1.453125,
29
+ "learning_rate": 0.001,
30
+ "loss": 22.35637664794922,
31
+ "step": 60
32
+ },
33
+ {
34
+ "epoch": 0.043138312213534646,
35
+ "grad_norm": 3.34375,
36
+ "learning_rate": 0.001,
37
+ "loss": 20.596783447265626,
38
+ "step": 80
39
+ },
40
+ {
41
+ "epoch": 0.05392289026691831,
42
+ "grad_norm": 1.9140625,
43
+ "learning_rate": 0.001,
44
+ "loss": 18.65278015136719,
45
+ "step": 100
46
+ },
47
+ {
48
+ "epoch": 0.06470746832030197,
49
+ "grad_norm": 3.109375,
50
+ "learning_rate": 0.001,
51
+ "loss": 17.364892578125,
52
+ "step": 120
53
+ },
54
+ {
55
+ "epoch": 0.07549204637368563,
56
+ "grad_norm": 1.6015625,
57
+ "learning_rate": 0.001,
58
+ "loss": 16.395790100097656,
59
+ "step": 140
60
+ },
61
+ {
62
+ "epoch": 0.08627662442706929,
63
+ "grad_norm": 1.5234375,
64
+ "learning_rate": 0.001,
65
+ "loss": 15.567823791503907,
66
+ "step": 160
67
+ },
68
+ {
69
+ "epoch": 0.09706120248045295,
70
+ "grad_norm": 1.703125,
71
+ "learning_rate": 0.001,
72
+ "loss": 14.909393310546875,
73
+ "step": 180
74
+ },
75
+ {
76
+ "epoch": 0.10784578053383662,
77
+ "grad_norm": 2.4375,
78
+ "learning_rate": 0.001,
79
+ "loss": 14.338616943359375,
80
+ "step": 200
81
+ },
82
+ {
83
+ "epoch": 0.11863035858722028,
84
+ "grad_norm": 3.0625,
85
+ "learning_rate": 0.001,
86
+ "loss": 13.839776611328125,
87
+ "step": 220
88
+ },
89
+ {
90
+ "epoch": 0.12941493664060394,
91
+ "grad_norm": 1.5703125,
92
+ "learning_rate": 0.001,
93
+ "loss": 13.418243408203125,
94
+ "step": 240
95
+ },
96
+ {
97
+ "epoch": 0.1401995146939876,
98
+ "grad_norm": 1.96875,
99
+ "learning_rate": 0.001,
100
+ "loss": 12.980410766601562,
101
+ "step": 260
102
+ },
103
+ {
104
+ "epoch": 0.15098409274737126,
105
+ "grad_norm": 2.578125,
106
+ "learning_rate": 0.001,
107
+ "loss": 12.627825164794922,
108
+ "step": 280
109
+ },
110
+ {
111
+ "epoch": 0.1617686708007549,
112
+ "grad_norm": 2.078125,
113
+ "learning_rate": 0.001,
114
+ "loss": 12.301615142822266,
115
+ "step": 300
116
+ },
117
+ {
118
+ "epoch": 0.17255324885413859,
119
+ "grad_norm": 2.09375,
120
+ "learning_rate": 0.001,
121
+ "loss": 11.97824478149414,
122
+ "step": 320
123
+ },
124
+ {
125
+ "epoch": 0.18333782690752223,
126
+ "grad_norm": 2.171875,
127
+ "learning_rate": 0.001,
128
+ "loss": 11.688941955566406,
129
+ "step": 340
130
+ },
131
+ {
132
+ "epoch": 0.1941224049609059,
133
+ "grad_norm": 2.125,
134
+ "learning_rate": 0.001,
135
+ "loss": 11.42403335571289,
136
+ "step": 360
137
+ },
138
+ {
139
+ "epoch": 0.20490698301428956,
140
+ "grad_norm": 2.234375,
141
+ "learning_rate": 0.001,
142
+ "loss": 11.158559417724609,
143
+ "step": 380
144
+ },
145
+ {
146
+ "epoch": 0.21569156106767323,
147
+ "grad_norm": 2.453125,
148
+ "learning_rate": 0.001,
149
+ "loss": 10.936893463134766,
150
+ "step": 400
151
+ },
152
+ {
153
+ "epoch": 0.22647613912105688,
154
+ "grad_norm": 2.078125,
155
+ "learning_rate": 0.001,
156
+ "loss": 10.734546661376953,
157
+ "step": 420
158
+ },
159
+ {
160
+ "epoch": 0.23726071717444056,
161
+ "grad_norm": 2.71875,
162
+ "learning_rate": 0.001,
163
+ "loss": 10.598049926757813,
164
+ "step": 440
165
+ },
166
+ {
167
+ "epoch": 0.2480452952278242,
168
+ "grad_norm": 2.515625,
169
+ "learning_rate": 0.001,
170
+ "loss": 10.398246002197265,
171
+ "step": 460
172
+ },
173
+ {
174
+ "epoch": 0.2588298732812079,
175
+ "grad_norm": 2.390625,
176
+ "learning_rate": 0.001,
177
+ "loss": 10.239346313476563,
178
+ "step": 480
179
+ },
180
+ {
181
+ "epoch": 0.26961445133459155,
182
+ "grad_norm": 2.25,
183
+ "learning_rate": 0.001,
184
+ "loss": 10.090744018554688,
185
+ "step": 500
186
+ }
187
+ ],
188
+ "logging_steps": 20,
189
+ "max_steps": 1000,
190
+ "num_input_tokens_seen": 0,
191
+ "num_train_epochs": 1,
192
+ "save_steps": 100,
193
+ "stateful_callbacks": {
194
+ "TrainerControl": {
195
+ "args": {
196
+ "should_epoch_stop": false,
197
+ "should_evaluate": false,
198
+ "should_log": false,
199
+ "should_save": true,
200
+ "should_training_stop": false
201
+ },
202
+ "attributes": {}
203
+ }
204
+ },
205
+ "total_flos": 2710157918208000.0,
206
+ "train_batch_size": 128,
207
+ "trial_name": null,
208
+ "trial_params": null
209
+ }
zain/Activation/out/glu-waleed10-21L_run/checkpoint-500/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d4a5fb74d92c2a824150af3ee3051d6e747acb606e30ff7dc799c234f6c87bd3
3
+ size 4920
zain/Activation/out/glu-waleed10-21L_run/checkpoint-600/config.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation": "waleed10",
3
+ "architectures": [
4
+ "TinyLlamaForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 1,
9
+ "dtype": "bfloat16",
10
+ "eos_token_id": 2,
11
+ "head_dim": 32,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 128,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 256,
16
+ "max_position_embeddings": 512,
17
+ "mlp_bias": false,
18
+ "mlp_type": "glu",
19
+ "model_type": "tiny_llama",
20
+ "num_attention_heads": 4,
21
+ "num_hidden_layers": 21,
22
+ "num_key_value_heads": 4,
23
+ "pad_token_id": 0,
24
+ "pretraining_tp": 1,
25
+ "rms_norm_eps": 1e-06,
26
+ "rope_parameters": {
27
+ "rope_theta": 10000.0,
28
+ "rope_type": "default"
29
+ },
30
+ "tie_word_embeddings": true,
31
+ "tokenizer_name": "w-ahmad/tiny-stories-tokenizer",
32
+ "transformers_version": "5.16.0.dev0",
33
+ "use_cache": false,
34
+ "vocab_size": 4096,
35
+ "waleed_beta": 10.0
36
+ }
zain/Activation/out/glu-waleed10-21L_run/checkpoint-600/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bcc71bb7f55325ed8c0ef097e9f17ca234fe2c331074f84d93746ae31ce7181e
3
+ size 7961528
zain/Activation/out/glu-waleed10-21L_run/checkpoint-600/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e2e352e0044d3cf63d7411c6dc38cf7b58373581524cf90d75327a2a564f4915
3
+ size 16041274
zain/Activation/out/glu-waleed10-21L_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-waleed10-21L_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-waleed10-21L_run/checkpoint-600/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
zain/Activation/out/glu-waleed10-21L_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-waleed10-21L_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": 28.876385498046876,
17
+ "step": 20
18
+ },
19
+ {
20
+ "epoch": 0.021569156106767323,
21
+ "grad_norm": 1.3984375,
22
+ "learning_rate": 0.001,
23
+ "loss": 24.010443115234374,
24
+ "step": 40
25
+ },
26
+ {
27
+ "epoch": 0.032353734160150985,
28
+ "grad_norm": 1.453125,
29
+ "learning_rate": 0.001,
30
+ "loss": 22.35637664794922,
31
+ "step": 60
32
+ },
33
+ {
34
+ "epoch": 0.043138312213534646,
35
+ "grad_norm": 3.34375,
36
+ "learning_rate": 0.001,
37
+ "loss": 20.596783447265626,
38
+ "step": 80
39
+ },
40
+ {
41
+ "epoch": 0.05392289026691831,
42
+ "grad_norm": 1.9140625,
43
+ "learning_rate": 0.001,
44
+ "loss": 18.65278015136719,
45
+ "step": 100
46
+ },
47
+ {
48
+ "epoch": 0.06470746832030197,
49
+ "grad_norm": 3.109375,
50
+ "learning_rate": 0.001,
51
+ "loss": 17.364892578125,
52
+ "step": 120
53
+ },
54
+ {
55
+ "epoch": 0.07549204637368563,
56
+ "grad_norm": 1.6015625,
57
+ "learning_rate": 0.001,
58
+ "loss": 16.395790100097656,
59
+ "step": 140
60
+ },
61
+ {
62
+ "epoch": 0.08627662442706929,
63
+ "grad_norm": 1.5234375,
64
+ "learning_rate": 0.001,
65
+ "loss": 15.567823791503907,
66
+ "step": 160
67
+ },
68
+ {
69
+ "epoch": 0.09706120248045295,
70
+ "grad_norm": 1.703125,
71
+ "learning_rate": 0.001,
72
+ "loss": 14.909393310546875,
73
+ "step": 180
74
+ },
75
+ {
76
+ "epoch": 0.10784578053383662,
77
+ "grad_norm": 2.4375,
78
+ "learning_rate": 0.001,
79
+ "loss": 14.338616943359375,
80
+ "step": 200
81
+ },
82
+ {
83
+ "epoch": 0.11863035858722028,
84
+ "grad_norm": 3.0625,
85
+ "learning_rate": 0.001,
86
+ "loss": 13.839776611328125,
87
+ "step": 220
88
+ },
89
+ {
90
+ "epoch": 0.12941493664060394,
91
+ "grad_norm": 1.5703125,
92
+ "learning_rate": 0.001,
93
+ "loss": 13.418243408203125,
94
+ "step": 240
95
+ },
96
+ {
97
+ "epoch": 0.1401995146939876,
98
+ "grad_norm": 1.96875,
99
+ "learning_rate": 0.001,
100
+ "loss": 12.980410766601562,
101
+ "step": 260
102
+ },
103
+ {
104
+ "epoch": 0.15098409274737126,
105
+ "grad_norm": 2.578125,
106
+ "learning_rate": 0.001,
107
+ "loss": 12.627825164794922,
108
+ "step": 280
109
+ },
110
+ {
111
+ "epoch": 0.1617686708007549,
112
+ "grad_norm": 2.078125,
113
+ "learning_rate": 0.001,
114
+ "loss": 12.301615142822266,
115
+ "step": 300
116
+ },
117
+ {
118
+ "epoch": 0.17255324885413859,
119
+ "grad_norm": 2.09375,
120
+ "learning_rate": 0.001,
121
+ "loss": 11.97824478149414,
122
+ "step": 320
123
+ },
124
+ {
125
+ "epoch": 0.18333782690752223,
126
+ "grad_norm": 2.171875,
127
+ "learning_rate": 0.001,
128
+ "loss": 11.688941955566406,
129
+ "step": 340
130
+ },
131
+ {
132
+ "epoch": 0.1941224049609059,
133
+ "grad_norm": 2.125,
134
+ "learning_rate": 0.001,
135
+ "loss": 11.42403335571289,
136
+ "step": 360
137
+ },
138
+ {
139
+ "epoch": 0.20490698301428956,
140
+ "grad_norm": 2.234375,
141
+ "learning_rate": 0.001,
142
+ "loss": 11.158559417724609,
143
+ "step": 380
144
+ },
145
+ {
146
+ "epoch": 0.21569156106767323,
147
+ "grad_norm": 2.453125,
148
+ "learning_rate": 0.001,
149
+ "loss": 10.936893463134766,
150
+ "step": 400
151
+ },
152
+ {
153
+ "epoch": 0.22647613912105688,
154
+ "grad_norm": 2.078125,
155
+ "learning_rate": 0.001,
156
+ "loss": 10.734546661376953,
157
+ "step": 420
158
+ },
159
+ {
160
+ "epoch": 0.23726071717444056,
161
+ "grad_norm": 2.71875,
162
+ "learning_rate": 0.001,
163
+ "loss": 10.598049926757813,
164
+ "step": 440
165
+ },
166
+ {
167
+ "epoch": 0.2480452952278242,
168
+ "grad_norm": 2.515625,
169
+ "learning_rate": 0.001,
170
+ "loss": 10.398246002197265,
171
+ "step": 460
172
+ },
173
+ {
174
+ "epoch": 0.2588298732812079,
175
+ "grad_norm": 2.390625,
176
+ "learning_rate": 0.001,
177
+ "loss": 10.239346313476563,
178
+ "step": 480
179
+ },
180
+ {
181
+ "epoch": 0.26961445133459155,
182
+ "grad_norm": 2.25,
183
+ "learning_rate": 0.001,
184
+ "loss": 10.090744018554688,
185
+ "step": 500
186
+ },
187
+ {
188
+ "epoch": 0.2803990293879752,
189
+ "grad_norm": 2.21875,
190
+ "learning_rate": 0.001,
191
+ "loss": 9.968287658691406,
192
+ "step": 520
193
+ },
194
+ {
195
+ "epoch": 0.29118360744135885,
196
+ "grad_norm": 2.015625,
197
+ "learning_rate": 0.001,
198
+ "loss": 9.856298828125,
199
+ "step": 540
200
+ },
201
+ {
202
+ "epoch": 0.3019681854947425,
203
+ "grad_norm": 2.4375,
204
+ "learning_rate": 0.001,
205
+ "loss": 9.739572906494141,
206
+ "step": 560
207
+ },
208
+ {
209
+ "epoch": 0.3127527635481262,
210
+ "grad_norm": 1.9765625,
211
+ "learning_rate": 0.001,
212
+ "loss": 9.626911163330078,
213
+ "step": 580
214
+ },
215
+ {
216
+ "epoch": 0.3235373416015098,
217
+ "grad_norm": 2.109375,
218
+ "learning_rate": 0.001,
219
+ "loss": 9.514720153808593,
220
+ "step": 600
221
+ }
222
+ ],
223
+ "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-waleed10-21L_run/checkpoint-600/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d4a5fb74d92c2a824150af3ee3051d6e747acb606e30ff7dc799c234f6c87bd3
3
+ size 4920
zain/Activation/out/glu-waleed10-21L_run/checkpoint-700/config.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation": "waleed10",
3
+ "architectures": [
4
+ "TinyLlamaForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 1,
9
+ "dtype": "bfloat16",
10
+ "eos_token_id": 2,
11
+ "head_dim": 32,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 128,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 256,
16
+ "max_position_embeddings": 512,
17
+ "mlp_bias": false,
18
+ "mlp_type": "glu",
19
+ "model_type": "tiny_llama",
20
+ "num_attention_heads": 4,
21
+ "num_hidden_layers": 21,
22
+ "num_key_value_heads": 4,
23
+ "pad_token_id": 0,
24
+ "pretraining_tp": 1,
25
+ "rms_norm_eps": 1e-06,
26
+ "rope_parameters": {
27
+ "rope_theta": 10000.0,
28
+ "rope_type": "default"
29
+ },
30
+ "tie_word_embeddings": true,
31
+ "tokenizer_name": "w-ahmad/tiny-stories-tokenizer",
32
+ "transformers_version": "5.16.0.dev0",
33
+ "use_cache": false,
34
+ "vocab_size": 4096,
35
+ "waleed_beta": 10.0
36
+ }
zain/Activation/out/glu-waleed10-21L_run/checkpoint-700/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:134139ee9fbc497795719c8ba01141109831a3a65e2b119f052f80ed8d8b9e27
3
+ size 7961528
zain/Activation/out/glu-waleed10-21L_run/checkpoint-700/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b906eeacaadd609e1c6d8aa32bfca24bd6e457b340be414612ed3b9b19dad560
3
+ size 16041274
zain/Activation/out/glu-waleed10-21L_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-waleed10-21L_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-waleed10-21L_run/checkpoint-700/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
zain/Activation/out/glu-waleed10-21L_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-waleed10-21L_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": 28.876385498046876,
17
+ "step": 20
18
+ },
19
+ {
20
+ "epoch": 0.021569156106767323,
21
+ "grad_norm": 1.3984375,
22
+ "learning_rate": 0.001,
23
+ "loss": 24.010443115234374,
24
+ "step": 40
25
+ },
26
+ {
27
+ "epoch": 0.032353734160150985,
28
+ "grad_norm": 1.453125,
29
+ "learning_rate": 0.001,
30
+ "loss": 22.35637664794922,
31
+ "step": 60
32
+ },
33
+ {
34
+ "epoch": 0.043138312213534646,
35
+ "grad_norm": 3.34375,
36
+ "learning_rate": 0.001,
37
+ "loss": 20.596783447265626,
38
+ "step": 80
39
+ },
40
+ {
41
+ "epoch": 0.05392289026691831,
42
+ "grad_norm": 1.9140625,
43
+ "learning_rate": 0.001,
44
+ "loss": 18.65278015136719,
45
+ "step": 100
46
+ },
47
+ {
48
+ "epoch": 0.06470746832030197,
49
+ "grad_norm": 3.109375,
50
+ "learning_rate": 0.001,
51
+ "loss": 17.364892578125,
52
+ "step": 120
53
+ },
54
+ {
55
+ "epoch": 0.07549204637368563,
56
+ "grad_norm": 1.6015625,
57
+ "learning_rate": 0.001,
58
+ "loss": 16.395790100097656,
59
+ "step": 140
60
+ },
61
+ {
62
+ "epoch": 0.08627662442706929,
63
+ "grad_norm": 1.5234375,
64
+ "learning_rate": 0.001,
65
+ "loss": 15.567823791503907,
66
+ "step": 160
67
+ },
68
+ {
69
+ "epoch": 0.09706120248045295,
70
+ "grad_norm": 1.703125,
71
+ "learning_rate": 0.001,
72
+ "loss": 14.909393310546875,
73
+ "step": 180
74
+ },
75
+ {
76
+ "epoch": 0.10784578053383662,
77
+ "grad_norm": 2.4375,
78
+ "learning_rate": 0.001,
79
+ "loss": 14.338616943359375,
80
+ "step": 200
81
+ },
82
+ {
83
+ "epoch": 0.11863035858722028,
84
+ "grad_norm": 3.0625,
85
+ "learning_rate": 0.001,
86
+ "loss": 13.839776611328125,
87
+ "step": 220
88
+ },
89
+ {
90
+ "epoch": 0.12941493664060394,
91
+ "grad_norm": 1.5703125,
92
+ "learning_rate": 0.001,
93
+ "loss": 13.418243408203125,
94
+ "step": 240
95
+ },
96
+ {
97
+ "epoch": 0.1401995146939876,
98
+ "grad_norm": 1.96875,
99
+ "learning_rate": 0.001,
100
+ "loss": 12.980410766601562,
101
+ "step": 260
102
+ },
103
+ {
104
+ "epoch": 0.15098409274737126,
105
+ "grad_norm": 2.578125,
106
+ "learning_rate": 0.001,
107
+ "loss": 12.627825164794922,
108
+ "step": 280
109
+ },
110
+ {
111
+ "epoch": 0.1617686708007549,
112
+ "grad_norm": 2.078125,
113
+ "learning_rate": 0.001,
114
+ "loss": 12.301615142822266,
115
+ "step": 300
116
+ },
117
+ {
118
+ "epoch": 0.17255324885413859,
119
+ "grad_norm": 2.09375,
120
+ "learning_rate": 0.001,
121
+ "loss": 11.97824478149414,
122
+ "step": 320
123
+ },
124
+ {
125
+ "epoch": 0.18333782690752223,
126
+ "grad_norm": 2.171875,
127
+ "learning_rate": 0.001,
128
+ "loss": 11.688941955566406,
129
+ "step": 340
130
+ },
131
+ {
132
+ "epoch": 0.1941224049609059,
133
+ "grad_norm": 2.125,
134
+ "learning_rate": 0.001,
135
+ "loss": 11.42403335571289,
136
+ "step": 360
137
+ },
138
+ {
139
+ "epoch": 0.20490698301428956,
140
+ "grad_norm": 2.234375,
141
+ "learning_rate": 0.001,
142
+ "loss": 11.158559417724609,
143
+ "step": 380
144
+ },
145
+ {
146
+ "epoch": 0.21569156106767323,
147
+ "grad_norm": 2.453125,
148
+ "learning_rate": 0.001,
149
+ "loss": 10.936893463134766,
150
+ "step": 400
151
+ },
152
+ {
153
+ "epoch": 0.22647613912105688,
154
+ "grad_norm": 2.078125,
155
+ "learning_rate": 0.001,
156
+ "loss": 10.734546661376953,
157
+ "step": 420
158
+ },
159
+ {
160
+ "epoch": 0.23726071717444056,
161
+ "grad_norm": 2.71875,
162
+ "learning_rate": 0.001,
163
+ "loss": 10.598049926757813,
164
+ "step": 440
165
+ },
166
+ {
167
+ "epoch": 0.2480452952278242,
168
+ "grad_norm": 2.515625,
169
+ "learning_rate": 0.001,
170
+ "loss": 10.398246002197265,
171
+ "step": 460
172
+ },
173
+ {
174
+ "epoch": 0.2588298732812079,
175
+ "grad_norm": 2.390625,
176
+ "learning_rate": 0.001,
177
+ "loss": 10.239346313476563,
178
+ "step": 480
179
+ },
180
+ {
181
+ "epoch": 0.26961445133459155,
182
+ "grad_norm": 2.25,
183
+ "learning_rate": 0.001,
184
+ "loss": 10.090744018554688,
185
+ "step": 500
186
+ },
187
+ {
188
+ "epoch": 0.2803990293879752,
189
+ "grad_norm": 2.21875,
190
+ "learning_rate": 0.001,
191
+ "loss": 9.968287658691406,
192
+ "step": 520
193
+ },
194
+ {
195
+ "epoch": 0.29118360744135885,
196
+ "grad_norm": 2.015625,
197
+ "learning_rate": 0.001,
198
+ "loss": 9.856298828125,
199
+ "step": 540
200
+ },
201
+ {
202
+ "epoch": 0.3019681854947425,
203
+ "grad_norm": 2.4375,
204
+ "learning_rate": 0.001,
205
+ "loss": 9.739572906494141,
206
+ "step": 560
207
+ },
208
+ {
209
+ "epoch": 0.3127527635481262,
210
+ "grad_norm": 1.9765625,
211
+ "learning_rate": 0.001,
212
+ "loss": 9.626911163330078,
213
+ "step": 580
214
+ },
215
+ {
216
+ "epoch": 0.3235373416015098,
217
+ "grad_norm": 2.109375,
218
+ "learning_rate": 0.001,
219
+ "loss": 9.514720153808593,
220
+ "step": 600
221
+ },
222
+ {
223
+ "epoch": 0.3343219196548935,
224
+ "grad_norm": 2.203125,
225
+ "learning_rate": 0.001,
226
+ "loss": 9.425536346435546,
227
+ "step": 620
228
+ },
229
+ {
230
+ "epoch": 0.34510649770827717,
231
+ "grad_norm": 2.296875,
232
+ "learning_rate": 0.001,
233
+ "loss": 9.341368865966796,
234
+ "step": 640
235
+ },
236
+ {
237
+ "epoch": 0.35589107576166085,
238
+ "grad_norm": 1.8203125,
239
+ "learning_rate": 0.001,
240
+ "loss": 9.254900360107422,
241
+ "step": 660
242
+ },
243
+ {
244
+ "epoch": 0.36667565381504447,
245
+ "grad_norm": 2.484375,
246
+ "learning_rate": 0.001,
247
+ "loss": 9.180448913574219,
248
+ "step": 680
249
+ },
250
+ {
251
+ "epoch": 0.37746023186842814,
252
+ "grad_norm": 2.515625,
253
+ "learning_rate": 0.001,
254
+ "loss": 9.107149505615235,
255
+ "step": 700
256
+ }
257
+ ],
258
+ "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-waleed10-21L_run/checkpoint-700/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d4a5fb74d92c2a824150af3ee3051d6e747acb606e30ff7dc799c234f6c87bd3
3
+ size 4920
zain/Activation/out/glu-waleed10-21L_run/checkpoint-800/config.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation": "waleed10",
3
+ "architectures": [
4
+ "TinyLlamaForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 1,
9
+ "dtype": "bfloat16",
10
+ "eos_token_id": 2,
11
+ "head_dim": 32,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 128,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 256,
16
+ "max_position_embeddings": 512,
17
+ "mlp_bias": false,
18
+ "mlp_type": "glu",
19
+ "model_type": "tiny_llama",
20
+ "num_attention_heads": 4,
21
+ "num_hidden_layers": 21,
22
+ "num_key_value_heads": 4,
23
+ "pad_token_id": 0,
24
+ "pretraining_tp": 1,
25
+ "rms_norm_eps": 1e-06,
26
+ "rope_parameters": {
27
+ "rope_theta": 10000.0,
28
+ "rope_type": "default"
29
+ },
30
+ "tie_word_embeddings": true,
31
+ "tokenizer_name": "w-ahmad/tiny-stories-tokenizer",
32
+ "transformers_version": "5.16.0.dev0",
33
+ "use_cache": false,
34
+ "vocab_size": 4096,
35
+ "waleed_beta": 10.0
36
+ }
zain/Activation/out/glu-waleed10-21L_run/checkpoint-800/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4122c8573503f8824b91613ce59c179385377c68670a5b34075c99f03474ceb5
3
+ size 7961528
zain/Activation/out/glu-waleed10-21L_run/checkpoint-800/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:16d6815b740fa6715b22eda33b408e85829b753a790e2b2b42fda2887ac8c029
3
+ size 16041274
zain/Activation/out/glu-waleed10-21L_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-waleed10-21L_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-waleed10-21L_run/checkpoint-800/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
zain/Activation/out/glu-waleed10-21L_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-waleed10-21L_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": 28.876385498046876,
17
+ "step": 20
18
+ },
19
+ {
20
+ "epoch": 0.021569156106767323,
21
+ "grad_norm": 1.3984375,
22
+ "learning_rate": 0.001,
23
+ "loss": 24.010443115234374,
24
+ "step": 40
25
+ },
26
+ {
27
+ "epoch": 0.032353734160150985,
28
+ "grad_norm": 1.453125,
29
+ "learning_rate": 0.001,
30
+ "loss": 22.35637664794922,
31
+ "step": 60
32
+ },
33
+ {
34
+ "epoch": 0.043138312213534646,
35
+ "grad_norm": 3.34375,
36
+ "learning_rate": 0.001,
37
+ "loss": 20.596783447265626,
38
+ "step": 80
39
+ },
40
+ {
41
+ "epoch": 0.05392289026691831,
42
+ "grad_norm": 1.9140625,
43
+ "learning_rate": 0.001,
44
+ "loss": 18.65278015136719,
45
+ "step": 100
46
+ },
47
+ {
48
+ "epoch": 0.06470746832030197,
49
+ "grad_norm": 3.109375,
50
+ "learning_rate": 0.001,
51
+ "loss": 17.364892578125,
52
+ "step": 120
53
+ },
54
+ {
55
+ "epoch": 0.07549204637368563,
56
+ "grad_norm": 1.6015625,
57
+ "learning_rate": 0.001,
58
+ "loss": 16.395790100097656,
59
+ "step": 140
60
+ },
61
+ {
62
+ "epoch": 0.08627662442706929,
63
+ "grad_norm": 1.5234375,
64
+ "learning_rate": 0.001,
65
+ "loss": 15.567823791503907,
66
+ "step": 160
67
+ },
68
+ {
69
+ "epoch": 0.09706120248045295,
70
+ "grad_norm": 1.703125,
71
+ "learning_rate": 0.001,
72
+ "loss": 14.909393310546875,
73
+ "step": 180
74
+ },
75
+ {
76
+ "epoch": 0.10784578053383662,
77
+ "grad_norm": 2.4375,
78
+ "learning_rate": 0.001,
79
+ "loss": 14.338616943359375,
80
+ "step": 200
81
+ },
82
+ {
83
+ "epoch": 0.11863035858722028,
84
+ "grad_norm": 3.0625,
85
+ "learning_rate": 0.001,
86
+ "loss": 13.839776611328125,
87
+ "step": 220
88
+ },
89
+ {
90
+ "epoch": 0.12941493664060394,
91
+ "grad_norm": 1.5703125,
92
+ "learning_rate": 0.001,
93
+ "loss": 13.418243408203125,
94
+ "step": 240
95
+ },
96
+ {
97
+ "epoch": 0.1401995146939876,
98
+ "grad_norm": 1.96875,
99
+ "learning_rate": 0.001,
100
+ "loss": 12.980410766601562,
101
+ "step": 260
102
+ },
103
+ {
104
+ "epoch": 0.15098409274737126,
105
+ "grad_norm": 2.578125,
106
+ "learning_rate": 0.001,
107
+ "loss": 12.627825164794922,
108
+ "step": 280
109
+ },
110
+ {
111
+ "epoch": 0.1617686708007549,
112
+ "grad_norm": 2.078125,
113
+ "learning_rate": 0.001,
114
+ "loss": 12.301615142822266,
115
+ "step": 300
116
+ },
117
+ {
118
+ "epoch": 0.17255324885413859,
119
+ "grad_norm": 2.09375,
120
+ "learning_rate": 0.001,
121
+ "loss": 11.97824478149414,
122
+ "step": 320
123
+ },
124
+ {
125
+ "epoch": 0.18333782690752223,
126
+ "grad_norm": 2.171875,
127
+ "learning_rate": 0.001,
128
+ "loss": 11.688941955566406,
129
+ "step": 340
130
+ },
131
+ {
132
+ "epoch": 0.1941224049609059,
133
+ "grad_norm": 2.125,
134
+ "learning_rate": 0.001,
135
+ "loss": 11.42403335571289,
136
+ "step": 360
137
+ },
138
+ {
139
+ "epoch": 0.20490698301428956,
140
+ "grad_norm": 2.234375,
141
+ "learning_rate": 0.001,
142
+ "loss": 11.158559417724609,
143
+ "step": 380
144
+ },
145
+ {
146
+ "epoch": 0.21569156106767323,
147
+ "grad_norm": 2.453125,
148
+ "learning_rate": 0.001,
149
+ "loss": 10.936893463134766,
150
+ "step": 400
151
+ },
152
+ {
153
+ "epoch": 0.22647613912105688,
154
+ "grad_norm": 2.078125,
155
+ "learning_rate": 0.001,
156
+ "loss": 10.734546661376953,
157
+ "step": 420
158
+ },
159
+ {
160
+ "epoch": 0.23726071717444056,
161
+ "grad_norm": 2.71875,
162
+ "learning_rate": 0.001,
163
+ "loss": 10.598049926757813,
164
+ "step": 440
165
+ },
166
+ {
167
+ "epoch": 0.2480452952278242,
168
+ "grad_norm": 2.515625,
169
+ "learning_rate": 0.001,
170
+ "loss": 10.398246002197265,
171
+ "step": 460
172
+ },
173
+ {
174
+ "epoch": 0.2588298732812079,
175
+ "grad_norm": 2.390625,
176
+ "learning_rate": 0.001,
177
+ "loss": 10.239346313476563,
178
+ "step": 480
179
+ },
180
+ {
181
+ "epoch": 0.26961445133459155,
182
+ "grad_norm": 2.25,
183
+ "learning_rate": 0.001,
184
+ "loss": 10.090744018554688,
185
+ "step": 500
186
+ },
187
+ {
188
+ "epoch": 0.2803990293879752,
189
+ "grad_norm": 2.21875,
190
+ "learning_rate": 0.001,
191
+ "loss": 9.968287658691406,
192
+ "step": 520
193
+ },
194
+ {
195
+ "epoch": 0.29118360744135885,
196
+ "grad_norm": 2.015625,
197
+ "learning_rate": 0.001,
198
+ "loss": 9.856298828125,
199
+ "step": 540
200
+ },
201
+ {
202
+ "epoch": 0.3019681854947425,
203
+ "grad_norm": 2.4375,
204
+ "learning_rate": 0.001,
205
+ "loss": 9.739572906494141,
206
+ "step": 560
207
+ },
208
+ {
209
+ "epoch": 0.3127527635481262,
210
+ "grad_norm": 1.9765625,
211
+ "learning_rate": 0.001,
212
+ "loss": 9.626911163330078,
213
+ "step": 580
214
+ },
215
+ {
216
+ "epoch": 0.3235373416015098,
217
+ "grad_norm": 2.109375,
218
+ "learning_rate": 0.001,
219
+ "loss": 9.514720153808593,
220
+ "step": 600
221
+ },
222
+ {
223
+ "epoch": 0.3343219196548935,
224
+ "grad_norm": 2.203125,
225
+ "learning_rate": 0.001,
226
+ "loss": 9.425536346435546,
227
+ "step": 620
228
+ },
229
+ {
230
+ "epoch": 0.34510649770827717,
231
+ "grad_norm": 2.296875,
232
+ "learning_rate": 0.001,
233
+ "loss": 9.341368865966796,
234
+ "step": 640
235
+ },
236
+ {
237
+ "epoch": 0.35589107576166085,
238
+ "grad_norm": 1.8203125,
239
+ "learning_rate": 0.001,
240
+ "loss": 9.254900360107422,
241
+ "step": 660
242
+ },
243
+ {
244
+ "epoch": 0.36667565381504447,
245
+ "grad_norm": 2.484375,
246
+ "learning_rate": 0.001,
247
+ "loss": 9.180448913574219,
248
+ "step": 680
249
+ },
250
+ {
251
+ "epoch": 0.37746023186842814,
252
+ "grad_norm": 2.515625,
253
+ "learning_rate": 0.001,
254
+ "loss": 9.107149505615235,
255
+ "step": 700
256
+ },
257
+ {
258
+ "epoch": 0.3882448099218118,
259
+ "grad_norm": 2.0,
260
+ "learning_rate": 0.001,
261
+ "loss": 9.010431671142578,
262
+ "step": 720
263
+ },
264
+ {
265
+ "epoch": 0.3990293879751955,
266
+ "grad_norm": 2.15625,
267
+ "learning_rate": 0.001,
268
+ "loss": 8.976707458496094,
269
+ "step": 740
270
+ },
271
+ {
272
+ "epoch": 0.4098139660285791,
273
+ "grad_norm": 1.9609375,
274
+ "learning_rate": 0.001,
275
+ "loss": 8.91417999267578,
276
+ "step": 760
277
+ },
278
+ {
279
+ "epoch": 0.4205985440819628,
280
+ "grad_norm": 1.8828125,
281
+ "learning_rate": 0.001,
282
+ "loss": 8.871702575683594,
283
+ "step": 780
284
+ },
285
+ {
286
+ "epoch": 0.43138312213534646,
287
+ "grad_norm": 2.0,
288
+ "learning_rate": 0.001,
289
+ "loss": 8.769727325439453,
290
+ "step": 800
291
+ }
292
+ ],
293
+ "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-waleed10-21L_run/checkpoint-800/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d4a5fb74d92c2a824150af3ee3051d6e747acb606e30ff7dc799c234f6c87bd3
3
+ size 4920
zain/Activation/out/glu-waleed10-21L_run/checkpoint-900/config.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation": "waleed10",
3
+ "architectures": [
4
+ "TinyLlamaForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 1,
9
+ "dtype": "bfloat16",
10
+ "eos_token_id": 2,
11
+ "head_dim": 32,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 128,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 256,
16
+ "max_position_embeddings": 512,
17
+ "mlp_bias": false,
18
+ "mlp_type": "glu",
19
+ "model_type": "tiny_llama",
20
+ "num_attention_heads": 4,
21
+ "num_hidden_layers": 21,
22
+ "num_key_value_heads": 4,
23
+ "pad_token_id": 0,
24
+ "pretraining_tp": 1,
25
+ "rms_norm_eps": 1e-06,
26
+ "rope_parameters": {
27
+ "rope_theta": 10000.0,
28
+ "rope_type": "default"
29
+ },
30
+ "tie_word_embeddings": true,
31
+ "tokenizer_name": "w-ahmad/tiny-stories-tokenizer",
32
+ "transformers_version": "5.16.0.dev0",
33
+ "use_cache": false,
34
+ "vocab_size": 4096,
35
+ "waleed_beta": 10.0
36
+ }
zain/Activation/out/glu-waleed10-21L_run/checkpoint-900/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:784c0a0b167617ae4036a6cc4bac84dff5af2404b10ed0a7645038e35eba332a
3
+ size 7961528
zain/Activation/out/glu-waleed10-21L_run/checkpoint-900/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:63870f4412169ed2e09cefc397d1f32694bcdd9993d891709d8149c8b006665c
3
+ size 16041274
zain/Activation/out/glu-waleed10-21L_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