Hugging Face's logo

w-ahmad
/
STRESS-TEST2

Safetensors
Model card Files
xet
 Community
w-ahmad commited on
Commit
4ee840d
·
verified ·
1 Parent(s): 6651ae4

Auto upload zain 2026-08-14T23:07:46.851605 (part 7)

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/mlp-waleed10-21L_run/checkpoint-100/config.json +36 -0
  2. zain/Activation/out/mlp-waleed10-21L_run/checkpoint-100/model.safetensors +3 -0
  3. zain/Activation/out/mlp-waleed10-21L_run/checkpoint-100/optimizer.pt +3 -0
  4. zain/Activation/out/mlp-waleed10-21L_run/checkpoint-100/rng_state.pth +3 -0
  5. zain/Activation/out/mlp-waleed10-21L_run/checkpoint-100/scheduler.pt +3 -0
  6. zain/Activation/out/mlp-waleed10-21L_run/checkpoint-100/tokenizer.json +0 -0
  7. zain/Activation/out/mlp-waleed10-21L_run/checkpoint-100/tokenizer_config.json +13 -0
  8. zain/Activation/out/mlp-waleed10-21L_run/checkpoint-100/trainer_state.json +69 -0
  9. zain/Activation/out/mlp-waleed10-21L_run/checkpoint-100/training_args.bin +3 -0
  10. zain/Activation/out/mlp-waleed10-21L_run/checkpoint-1000/config.json +36 -0
  11. zain/Activation/out/mlp-waleed10-21L_run/checkpoint-1000/model.safetensors +3 -0
  12. zain/Activation/out/mlp-waleed10-21L_run/checkpoint-1000/optimizer.pt +3 -0
  13. zain/Activation/out/mlp-waleed10-21L_run/checkpoint-1000/rng_state.pth +3 -0
  14. zain/Activation/out/mlp-waleed10-21L_run/checkpoint-1000/scheduler.pt +3 -0
  15. zain/Activation/out/mlp-waleed10-21L_run/checkpoint-1000/tokenizer.json +0 -0
  16. zain/Activation/out/mlp-waleed10-21L_run/checkpoint-1000/tokenizer_config.json +13 -0
  17. zain/Activation/out/mlp-waleed10-21L_run/checkpoint-1000/trainer_state.json +392 -0
  18. zain/Activation/out/mlp-waleed10-21L_run/checkpoint-1000/training_args.bin +3 -0
  19. zain/Activation/out/mlp-waleed10-21L_run/checkpoint-200/config.json +36 -0
  20. zain/Activation/out/mlp-waleed10-21L_run/checkpoint-200/model.safetensors +3 -0
  21. zain/Activation/out/mlp-waleed10-21L_run/checkpoint-200/optimizer.pt +3 -0
  22. zain/Activation/out/mlp-waleed10-21L_run/checkpoint-200/rng_state.pth +3 -0
  23. zain/Activation/out/mlp-waleed10-21L_run/checkpoint-200/scheduler.pt +3 -0
  24. zain/Activation/out/mlp-waleed10-21L_run/checkpoint-200/tokenizer.json +0 -0
  25. zain/Activation/out/mlp-waleed10-21L_run/checkpoint-200/tokenizer_config.json +13 -0
  26. zain/Activation/out/mlp-waleed10-21L_run/checkpoint-200/trainer_state.json +104 -0
  27. zain/Activation/out/mlp-waleed10-21L_run/checkpoint-200/training_args.bin +3 -0
  28. zain/Activation/out/mlp-waleed10-21L_run/checkpoint-300/config.json +36 -0
  29. zain/Activation/out/mlp-waleed10-21L_run/checkpoint-300/model.safetensors +3 -0
  30. zain/Activation/out/mlp-waleed10-21L_run/checkpoint-300/optimizer.pt +3 -0
  31. zain/Activation/out/mlp-waleed10-21L_run/checkpoint-300/rng_state.pth +3 -0
  32. zain/Activation/out/mlp-waleed10-21L_run/checkpoint-300/scheduler.pt +3 -0
  33. zain/Activation/out/mlp-waleed10-21L_run/checkpoint-300/tokenizer.json +0 -0
  34. zain/Activation/out/mlp-waleed10-21L_run/checkpoint-300/tokenizer_config.json +13 -0
  35. zain/Activation/out/mlp-waleed10-21L_run/checkpoint-300/trainer_state.json +139 -0
  36. zain/Activation/out/mlp-waleed10-21L_run/checkpoint-300/training_args.bin +3 -0
  37. zain/Activation/out/mlp-waleed10-21L_run/checkpoint-400/config.json +36 -0
  38. zain/Activation/out/mlp-waleed10-21L_run/checkpoint-400/model.safetensors +3 -0
  39. zain/Activation/out/mlp-waleed10-21L_run/checkpoint-400/optimizer.pt +3 -0
  40. zain/Activation/out/mlp-waleed10-21L_run/checkpoint-400/rng_state.pth +3 -0
  41. zain/Activation/out/mlp-waleed10-21L_run/checkpoint-400/scheduler.pt +3 -0
  42. zain/Activation/out/mlp-waleed10-21L_run/checkpoint-400/tokenizer.json +0 -0
  43. zain/Activation/out/mlp-waleed10-21L_run/checkpoint-400/tokenizer_config.json +13 -0
  44. zain/Activation/out/mlp-waleed10-21L_run/checkpoint-400/trainer_state.json +174 -0
  45. zain/Activation/out/mlp-waleed10-21L_run/checkpoint-400/training_args.bin +3 -0
  46. zain/Activation/out/mlp-waleed10-21L_run/checkpoint-500/config.json +36 -0
  47. zain/Activation/out/mlp-waleed10-21L_run/checkpoint-500/model.safetensors +3 -0
  48. zain/Activation/out/mlp-waleed10-21L_run/checkpoint-500/optimizer.pt +3 -0
  49. zain/Activation/out/mlp-waleed10-21L_run/checkpoint-500/rng_state.pth +3 -0
  50. zain/Activation/out/mlp-waleed10-21L_run/checkpoint-500/scheduler.pt +3 -0
zain/Activation/out/mlp-waleed10-21L_run/checkpoint-100/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": "mlp",
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/mlp-waleed10-21L_run/checkpoint-100/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b7459f635cb043dbf789935d75e6815eccfc28a97297bd54183df6ebe940da9c
3
+ size 7959296
zain/Activation/out/mlp-waleed10-21L_run/checkpoint-100/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2b8de4350d7698966f5c9eb811e52aa3142d0832cab02dce0a2005f6683790d7
3
+ size 16023738
zain/Activation/out/mlp-waleed10-21L_run/checkpoint-100/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/mlp-waleed10-21L_run/checkpoint-100/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:906745ab5f61e73e8c1ba850ef3b839e2766eba889b49c843836266988895165
3
+ size 1064
zain/Activation/out/mlp-waleed10-21L_run/checkpoint-100/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
zain/Activation/out/mlp-waleed10-21L_run/checkpoint-100/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/mlp-waleed10-21L_run/checkpoint-100/trainer_state.json ADDED
@@ -0,0 +1,69 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 0.05392289026691831,
6
+ "eval_steps": 60000,
7
+ "global_step": 100,
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": 4.21875,
15
+ "learning_rate": 0.001,
16
+ "loss": 29.101596069335937,
17
+ "step": 20
18
+ },
19
+ {
20
+ "epoch": 0.021569156106767323,
21
+ "grad_norm": 13.25,
22
+ "learning_rate": 0.001,
23
+ "loss": 24.024703979492188,
24
+ "step": 40
25
+ },
26
+ {
27
+ "epoch": 0.032353734160150985,
28
+ "grad_norm": 2.15625,
29
+ "learning_rate": 0.001,
30
+ "loss": 22.0672119140625,
31
+ "step": 60
32
+ },
33
+ {
34
+ "epoch": 0.043138312213534646,
35
+ "grad_norm": 2.984375,
36
+ "learning_rate": 0.001,
37
+ "loss": 20.730134582519533,
38
+ "step": 80
39
+ },
40
+ {
41
+ "epoch": 0.05392289026691831,
42
+ "grad_norm": 2.5625,
43
+ "learning_rate": 0.001,
44
+ "loss": 19.62140808105469,
45
+ "step": 100
46
+ }
47
+ ],
48
+ "logging_steps": 20,
49
+ "max_steps": 1000,
50
+ "num_input_tokens_seen": 0,
51
+ "num_train_epochs": 1,
52
+ "save_steps": 100,
53
+ "stateful_callbacks": {
54
+ "TrainerControl": {
55
+ "args": {
56
+ "should_epoch_stop": false,
57
+ "should_evaluate": false,
58
+ "should_log": false,
59
+ "should_save": true,
60
+ "should_training_stop": false
61
+ },
62
+ "attributes": {}
63
+ }
64
+ },
65
+ "total_flos": 542031583641600.0,
66
+ "train_batch_size": 128,
67
+ "trial_name": null,
68
+ "trial_params": null
69
+ }
zain/Activation/out/mlp-waleed10-21L_run/checkpoint-100/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1338c5cc45c605f6ad442767530c8d502dcc6ccd81ff99275bf324e95cbcf422
3
+ size 4920
zain/Activation/out/mlp-waleed10-21L_run/checkpoint-1000/config.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation": "waleed10",
3
+ "architectures": [
4
+ "TinyLlamaForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 1,
9
+ "dtype": "bfloat16",
10
+ "eos_token_id": 2,
11
+ "head_dim": 32,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 128,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 256,
16
+ "max_position_embeddings": 512,
17
+ "mlp_bias": false,
18
+ "mlp_type": "mlp",
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/mlp-waleed10-21L_run/checkpoint-1000/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:10aac04444e1ace3826f1af7d2cde0eaa919c75c6187d476daec0e9b5e3608de
3
+ size 7959296
zain/Activation/out/mlp-waleed10-21L_run/checkpoint-1000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3e161813b27c85c41a0e46134fd5543014e623eb216faf8988a067366a996405
3
+ size 16023738
zain/Activation/out/mlp-waleed10-21L_run/checkpoint-1000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2b66e3cc7c452b707ddac5caf0aa17618afb9bc1a0333600a22c4afb353f3165
3
+ size 14244
zain/Activation/out/mlp-waleed10-21L_run/checkpoint-1000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0fc3ca8af69cc9d003a05a139e7997960ab393fe31ae1d898b29038573704b4e
3
+ size 1064
zain/Activation/out/mlp-waleed10-21L_run/checkpoint-1000/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
zain/Activation/out/mlp-waleed10-21L_run/checkpoint-1000/tokenizer_config.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "backend": "tokenizers",
4
+ "bos_token": "<|endoftext|>",
5
+ "eos_token": "<|endoftext|>",
6
+ "errors": "replace",
7
+ "is_local": false,
8
+ "local_files_only": false,
9
+ "model_max_length": 1024,
10
+ "pad_token": "<|endoftext|>",
11
+ "tokenizer_class": "GPT2Tokenizer",
12
+ "unk_token": "<|endoftext|>"
13
+ }
zain/Activation/out/mlp-waleed10-21L_run/checkpoint-1000/trainer_state.json ADDED
@@ -0,0 +1,392 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 0.5392289026691831,
6
+ "eval_steps": 60000,
7
+ "global_step": 1000,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.010784578053383662,
14
+ "grad_norm": 4.21875,
15
+ "learning_rate": 0.001,
16
+ "loss": 29.101596069335937,
17
+ "step": 20
18
+ },
19
+ {
20
+ "epoch": 0.021569156106767323,
21
+ "grad_norm": 13.25,
22
+ "learning_rate": 0.001,
23
+ "loss": 24.024703979492188,
24
+ "step": 40
25
+ },
26
+ {
27
+ "epoch": 0.032353734160150985,
28
+ "grad_norm": 2.15625,
29
+ "learning_rate": 0.001,
30
+ "loss": 22.0672119140625,
31
+ "step": 60
32
+ },
33
+ {
34
+ "epoch": 0.043138312213534646,
35
+ "grad_norm": 2.984375,
36
+ "learning_rate": 0.001,
37
+ "loss": 20.730134582519533,
38
+ "step": 80
39
+ },
40
+ {
41
+ "epoch": 0.05392289026691831,
42
+ "grad_norm": 2.5625,
43
+ "learning_rate": 0.001,
44
+ "loss": 19.62140808105469,
45
+ "step": 100
46
+ },
47
+ {
48
+ "epoch": 0.06470746832030197,
49
+ "grad_norm": 3.5,
50
+ "learning_rate": 0.001,
51
+ "loss": 18.49439697265625,
52
+ "step": 120
53
+ },
54
+ {
55
+ "epoch": 0.07549204637368563,
56
+ "grad_norm": 2.328125,
57
+ "learning_rate": 0.001,
58
+ "loss": 17.532830810546876,
59
+ "step": 140
60
+ },
61
+ {
62
+ "epoch": 0.08627662442706929,
63
+ "grad_norm": 4.0,
64
+ "learning_rate": 0.001,
65
+ "loss": 16.738078308105468,
66
+ "step": 160
67
+ },
68
+ {
69
+ "epoch": 0.09706120248045295,
70
+ "grad_norm": 2.65625,
71
+ "learning_rate": 0.001,
72
+ "loss": 16.187765502929686,
73
+ "step": 180
74
+ },
75
+ {
76
+ "epoch": 0.10784578053383662,
77
+ "grad_norm": 2.046875,
78
+ "learning_rate": 0.001,
79
+ "loss": 15.680935668945313,
80
+ "step": 200
81
+ },
82
+ {
83
+ "epoch": 0.11863035858722028,
84
+ "grad_norm": 3.90625,
85
+ "learning_rate": 0.001,
86
+ "loss": 15.221630859375,
87
+ "step": 220
88
+ },
89
+ {
90
+ "epoch": 0.12941493664060394,
91
+ "grad_norm": 2.75,
92
+ "learning_rate": 0.001,
93
+ "loss": 14.924528503417969,
94
+ "step": 240
95
+ },
96
+ {
97
+ "epoch": 0.1401995146939876,
98
+ "grad_norm": 1.6328125,
99
+ "learning_rate": 0.001,
100
+ "loss": 14.54259796142578,
101
+ "step": 260
102
+ },
103
+ {
104
+ "epoch": 0.15098409274737126,
105
+ "grad_norm": 2.6875,
106
+ "learning_rate": 0.001,
107
+ "loss": 14.268138122558593,
108
+ "step": 280
109
+ },
110
+ {
111
+ "epoch": 0.1617686708007549,
112
+ "grad_norm": 3.15625,
113
+ "learning_rate": 0.001,
114
+ "loss": 14.015008544921875,
115
+ "step": 300
116
+ },
117
+ {
118
+ "epoch": 0.17255324885413859,
119
+ "grad_norm": 3.9375,
120
+ "learning_rate": 0.001,
121
+ "loss": 13.782272338867188,
122
+ "step": 320
123
+ },
124
+ {
125
+ "epoch": 0.18333782690752223,
126
+ "grad_norm": 2.90625,
127
+ "learning_rate": 0.001,
128
+ "loss": 13.597767639160157,
129
+ "step": 340
130
+ },
131
+ {
132
+ "epoch": 0.1941224049609059,
133
+ "grad_norm": 2.21875,
134
+ "learning_rate": 0.001,
135
+ "loss": 13.370980834960937,
136
+ "step": 360
137
+ },
138
+ {
139
+ "epoch": 0.20490698301428956,
140
+ "grad_norm": 3.796875,
141
+ "learning_rate": 0.001,
142
+ "loss": 13.148257446289062,
143
+ "step": 380
144
+ },
145
+ {
146
+ "epoch": 0.21569156106767323,
147
+ "grad_norm": 1.859375,
148
+ "learning_rate": 0.001,
149
+ "loss": 13.025798034667968,
150
+ "step": 400
151
+ },
152
+ {
153
+ "epoch": 0.22647613912105688,
154
+ "grad_norm": 2.765625,
155
+ "learning_rate": 0.001,
156
+ "loss": 12.835781860351563,
157
+ "step": 420
158
+ },
159
+ {
160
+ "epoch": 0.23726071717444056,
161
+ "grad_norm": 4.53125,
162
+ "learning_rate": 0.001,
163
+ "loss": 12.72701187133789,
164
+ "step": 440
165
+ },
166
+ {
167
+ "epoch": 0.2480452952278242,
168
+ "grad_norm": 3.03125,
169
+ "learning_rate": 0.001,
170
+ "loss": 12.553873443603516,
171
+ "step": 460
172
+ },
173
+ {
174
+ "epoch": 0.2588298732812079,
175
+ "grad_norm": 3.359375,
176
+ "learning_rate": 0.001,
177
+ "loss": 12.410502624511718,
178
+ "step": 480
179
+ },
180
+ {
181
+ "epoch": 0.26961445133459155,
182
+ "grad_norm": 3.609375,
183
+ "learning_rate": 0.001,
184
+ "loss": 12.281945037841798,
185
+ "step": 500
186
+ },
187
+ {
188
+ "epoch": 0.2803990293879752,
189
+ "grad_norm": 2.40625,
190
+ "learning_rate": 0.001,
191
+ "loss": 12.149514770507812,
192
+ "step": 520
193
+ },
194
+ {
195
+ "epoch": 0.29118360744135885,
196
+ "grad_norm": 3.0,
197
+ "learning_rate": 0.001,
198
+ "loss": 12.051705932617187,
199
+ "step": 540
200
+ },
201
+ {
202
+ "epoch": 0.3019681854947425,
203
+ "grad_norm": 2.546875,
204
+ "learning_rate": 0.001,
205
+ "loss": 11.956787872314454,
206
+ "step": 560
207
+ },
208
+ {
209
+ "epoch": 0.3127527635481262,
210
+ "grad_norm": 2.25,
211
+ "learning_rate": 0.001,
212
+ "loss": 11.817385864257812,
213
+ "step": 580
214
+ },
215
+ {
216
+ "epoch": 0.3235373416015098,
217
+ "grad_norm": 3.21875,
218
+ "learning_rate": 0.001,
219
+ "loss": 11.71741180419922,
220
+ "step": 600
221
+ },
222
+ {
223
+ "epoch": 0.3343219196548935,
224
+ "grad_norm": 2.4375,
225
+ "learning_rate": 0.001,
226
+ "loss": 11.652935791015626,
227
+ "step": 620
228
+ },
229
+ {
230
+ "epoch": 0.34510649770827717,
231
+ "grad_norm": 2.46875,
232
+ "learning_rate": 0.001,
233
+ "loss": 11.560204315185548,
234
+ "step": 640
235
+ },
236
+ {
237
+ "epoch": 0.35589107576166085,
238
+ "grad_norm": 2.5,
239
+ "learning_rate": 0.001,
240
+ "loss": 11.46844253540039,
241
+ "step": 660
242
+ },
243
+ {
244
+ "epoch": 0.36667565381504447,
245
+ "grad_norm": 2.90625,
246
+ "learning_rate": 0.001,
247
+ "loss": 11.36828155517578,
248
+ "step": 680
249
+ },
250
+ {
251
+ "epoch": 0.37746023186842814,
252
+ "grad_norm": 3.140625,
253
+ "learning_rate": 0.001,
254
+ "loss": 11.288950347900391,
255
+ "step": 700
256
+ },
257
+ {
258
+ "epoch": 0.3882448099218118,
259
+ "grad_norm": 2.96875,
260
+ "learning_rate": 0.001,
261
+ "loss": 11.199080657958984,
262
+ "step": 720
263
+ },
264
+ {
265
+ "epoch": 0.3990293879751955,
266
+ "grad_norm": 2.734375,
267
+ "learning_rate": 0.001,
268
+ "loss": 11.141081237792969,
269
+ "step": 740
270
+ },
271
+ {
272
+ "epoch": 0.4098139660285791,
273
+ "grad_norm": 2.21875,
274
+ "learning_rate": 0.001,
275
+ "loss": 11.086531066894532,
276
+ "step": 760
277
+ },
278
+ {
279
+ "epoch": 0.4205985440819628,
280
+ "grad_norm": 2.796875,
281
+ "learning_rate": 0.001,
282
+ "loss": 11.031404113769531,
283
+ "step": 780
284
+ },
285
+ {
286
+ "epoch": 0.43138312213534646,
287
+ "grad_norm": 3.359375,
288
+ "learning_rate": 0.001,
289
+ "loss": 10.9326171875,
290
+ "step": 800
291
+ },
292
+ {
293
+ "epoch": 0.44216770018873014,
294
+ "grad_norm": 2.46875,
295
+ "learning_rate": 0.001,
296
+ "loss": 10.880109405517578,
297
+ "step": 820
298
+ },
299
+ {
300
+ "epoch": 0.45295227824211376,
301
+ "grad_norm": 2.796875,
302
+ "learning_rate": 0.001,
303
+ "loss": 10.829965972900391,
304
+ "step": 840
305
+ },
306
+ {
307
+ "epoch": 0.46373685629549743,
308
+ "grad_norm": 2.46875,
309
+ "learning_rate": 0.001,
310
+ "loss": 10.754279327392577,
311
+ "step": 860
312
+ },
313
+ {
314
+ "epoch": 0.4745214343488811,
315
+ "grad_norm": 2.84375,
316
+ "learning_rate": 0.001,
317
+ "loss": 10.709523773193359,
318
+ "step": 880
319
+ },
320
+ {
321
+ "epoch": 0.4853060124022648,
322
+ "grad_norm": 2.75,
323
+ "learning_rate": 0.001,
324
+ "loss": 10.660047912597657,
325
+ "step": 900
326
+ },
327
+ {
328
+ "epoch": 0.4960905904556484,
329
+ "grad_norm": 2.421875,
330
+ "learning_rate": 0.001,
331
+ "loss": 10.584401702880859,
332
+ "step": 920
333
+ },
334
+ {
335
+ "epoch": 0.5068751685090321,
336
+ "grad_norm": 2.21875,
337
+ "learning_rate": 0.001,
338
+ "loss": 10.558298492431641,
339
+ "step": 940
340
+ },
341
+ {
342
+ "epoch": 0.5176597465624158,
343
+ "grad_norm": 2.328125,
344
+ "learning_rate": 0.001,
345
+ "loss": 10.5265380859375,
346
+ "step": 960
347
+ },
348
+ {
349
+ "epoch": 0.5284443246157994,
350
+ "grad_norm": 3.453125,
351
+ "learning_rate": 0.001,
352
+ "loss": 10.47811508178711,
353
+ "step": 980
354
+ },
355
+ {
356
+ "epoch": 0.5392289026691831,
357
+ "grad_norm": 2.375,
358
+ "learning_rate": 0.001,
359
+ "loss": 10.437238311767578,
360
+ "step": 1000
361
+ },
362
+ {
363
+ "epoch": 0.5392289026691831,
364
+ "eval_loss": 2.6014304161071777,
365
+ "eval_runtime": 9.308,
366
+ "eval_samples_per_second": 1023.53,
367
+ "eval_steps_per_second": 8.058,
368
+ "step": 1000
369
+ }
370
+ ],
371
+ "logging_steps": 20,
372
+ "max_steps": 1000,
373
+ "num_input_tokens_seen": 0,
374
+ "num_train_epochs": 1,
375
+ "save_steps": 100,
376
+ "stateful_callbacks": {
377
+ "TrainerControl": {
378
+ "args": {
379
+ "should_epoch_stop": false,
380
+ "should_evaluate": false,
381
+ "should_log": false,
382
+ "should_save": true,
383
+ "should_training_stop": true
384
+ },
385
+ "attributes": {}
386
+ }
387
+ },
388
+ "total_flos": 5420315836416000.0,
389
+ "train_batch_size": 128,
390
+ "trial_name": null,
391
+ "trial_params": null
392
+ }
zain/Activation/out/mlp-waleed10-21L_run/checkpoint-1000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1338c5cc45c605f6ad442767530c8d502dcc6ccd81ff99275bf324e95cbcf422
3
+ size 4920
zain/Activation/out/mlp-waleed10-21L_run/checkpoint-200/config.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation": "waleed10",
3
+ "architectures": [
4
+ "TinyLlamaForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 1,
9
+ "dtype": "bfloat16",
10
+ "eos_token_id": 2,
11
+ "head_dim": 32,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 128,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 256,
16
+ "max_position_embeddings": 512,
17
+ "mlp_bias": false,
18
+ "mlp_type": "mlp",
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/mlp-waleed10-21L_run/checkpoint-200/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ac951199daec0a0f05091b9797720dc576cfcf2d6558c7441779bdbedf0eda83
3
+ size 7959296
zain/Activation/out/mlp-waleed10-21L_run/checkpoint-200/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:45851c852670030798c69d3c78ded53de5f4563ba705b0af5803e1e6ba09334e
3
+ size 16023738
zain/Activation/out/mlp-waleed10-21L_run/checkpoint-200/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9d9cd6a0487226e5bd30d1846894c82af483733ab4381b75bae9c0745e05d405
3
+ size 14244
zain/Activation/out/mlp-waleed10-21L_run/checkpoint-200/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2d3b5a970fc92af42664e84ba686059eaaacfaa668691c0cf6d2afddda677461
3
+ size 1064
zain/Activation/out/mlp-waleed10-21L_run/checkpoint-200/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
zain/Activation/out/mlp-waleed10-21L_run/checkpoint-200/tokenizer_config.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "backend": "tokenizers",
4
+ "bos_token": "<|endoftext|>",
5
+ "eos_token": "<|endoftext|>",
6
+ "errors": "replace",
7
+ "is_local": false,
8
+ "local_files_only": false,
9
+ "model_max_length": 1024,
10
+ "pad_token": "<|endoftext|>",
11
+ "tokenizer_class": "GPT2Tokenizer",
12
+ "unk_token": "<|endoftext|>"
13
+ }
zain/Activation/out/mlp-waleed10-21L_run/checkpoint-200/trainer_state.json ADDED
@@ -0,0 +1,104 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 0.10784578053383662,
6
+ "eval_steps": 60000,
7
+ "global_step": 200,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.010784578053383662,
14
+ "grad_norm": 4.21875,
15
+ "learning_rate": 0.001,
16
+ "loss": 29.101596069335937,
17
+ "step": 20
18
+ },
19
+ {
20
+ "epoch": 0.021569156106767323,
21
+ "grad_norm": 13.25,
22
+ "learning_rate": 0.001,
23
+ "loss": 24.024703979492188,
24
+ "step": 40
25
+ },
26
+ {
27
+ "epoch": 0.032353734160150985,
28
+ "grad_norm": 2.15625,
29
+ "learning_rate": 0.001,
30
+ "loss": 22.0672119140625,
31
+ "step": 60
32
+ },
33
+ {
34
+ "epoch": 0.043138312213534646,
35
+ "grad_norm": 2.984375,
36
+ "learning_rate": 0.001,
37
+ "loss": 20.730134582519533,
38
+ "step": 80
39
+ },
40
+ {
41
+ "epoch": 0.05392289026691831,
42
+ "grad_norm": 2.5625,
43
+ "learning_rate": 0.001,
44
+ "loss": 19.62140808105469,
45
+ "step": 100
46
+ },
47
+ {
48
+ "epoch": 0.06470746832030197,
49
+ "grad_norm": 3.5,
50
+ "learning_rate": 0.001,
51
+ "loss": 18.49439697265625,
52
+ "step": 120
53
+ },
54
+ {
55
+ "epoch": 0.07549204637368563,
56
+ "grad_norm": 2.328125,
57
+ "learning_rate": 0.001,
58
+ "loss": 17.532830810546876,
59
+ "step": 140
60
+ },
61
+ {
62
+ "epoch": 0.08627662442706929,
63
+ "grad_norm": 4.0,
64
+ "learning_rate": 0.001,
65
+ "loss": 16.738078308105468,
66
+ "step": 160
67
+ },
68
+ {
69
+ "epoch": 0.09706120248045295,
70
+ "grad_norm": 2.65625,
71
+ "learning_rate": 0.001,
72
+ "loss": 16.187765502929686,
73
+ "step": 180
74
+ },
75
+ {
76
+ "epoch": 0.10784578053383662,
77
+ "grad_norm": 2.046875,
78
+ "learning_rate": 0.001,
79
+ "loss": 15.680935668945313,
80
+ "step": 200
81
+ }
82
+ ],
83
+ "logging_steps": 20,
84
+ "max_steps": 1000,
85
+ "num_input_tokens_seen": 0,
86
+ "num_train_epochs": 1,
87
+ "save_steps": 100,
88
+ "stateful_callbacks": {
89
+ "TrainerControl": {
90
+ "args": {
91
+ "should_epoch_stop": false,
92
+ "should_evaluate": false,
93
+ "should_log": false,
94
+ "should_save": true,
95
+ "should_training_stop": false
96
+ },
97
+ "attributes": {}
98
+ }
99
+ },
100
+ "total_flos": 1084063167283200.0,
101
+ "train_batch_size": 128,
102
+ "trial_name": null,
103
+ "trial_params": null
104
+ }
zain/Activation/out/mlp-waleed10-21L_run/checkpoint-200/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1338c5cc45c605f6ad442767530c8d502dcc6ccd81ff99275bf324e95cbcf422
3
+ size 4920
zain/Activation/out/mlp-waleed10-21L_run/checkpoint-300/config.json ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation": "waleed10",
3
+ "architectures": [
4
+ "TinyLlamaForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 1,
9
+ "dtype": "bfloat16",
10
+ "eos_token_id": 2,
11
+ "head_dim": 32,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 128,
14
+ "initializer_range": 0.02,
15
+ "intermediate_size": 256,
16
+ "max_position_embeddings": 512,
17
+ "mlp_bias": false,
18
+ "mlp_type": "mlp",
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/mlp-waleed10-21L_run/checkpoint-300/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:24fd84f4ab5b50de588ba0a6b3464bc4d8a3ba4171c79bd890a2807249eb6a7a
3
+ size 7959296
zain/Activation/out/mlp-waleed10-21L_run/checkpoint-300/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2fdd0bca894816ea0354eba8e2b900d5ed09bf55201695f7f3415e6343037a03
3
+ size 16023738
zain/Activation/out/mlp-waleed10-21L_run/checkpoint-300/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9d9cd6a0487226e5bd30d1846894c82af483733ab4381b75bae9c0745e05d405
3
+ size 14244
zain/Activation/out/mlp-waleed10-21L_run/checkpoint-300/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:76cd21f32221b90fb8288fd2428595ec8b5039fe2425086d67cacb5c8d814c9e
3
+ size 1064
zain/Activation/out/mlp-waleed10-21L_run/checkpoint-300/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
zain/Activation/out/mlp-waleed10-21L_run/checkpoint-300/tokenizer_config.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "backend": "tokenizers",
4
+ "bos_token": "<|endoftext|>",
5
+ "eos_token": "<|endoftext|>",
6
+ "errors": "replace",
7
+ "is_local": false,
8
+ "local_files_only": false,
9
+ "model_max_length": 1024,
10
+ "pad_token": "<|endoftext|>",
11
+ "tokenizer_class": "GPT2Tokenizer",
12
+ "unk_token": "<|endoftext|>"
13
+ }
zain/Activation/out/mlp-waleed10-21L_run/checkpoint-300/trainer_state.json ADDED
@@ -0,0 +1,139 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 0.1617686708007549,
6
+ "eval_steps": 60000,
7
+ "global_step": 300,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.010784578053383662,
14
+ "grad_norm": 4.21875,
15
+ "learning_rate": 0.001,
16
+ "loss": 29.101596069335937,
17
+ "step": 20
18
+ },
19
+ {
20
+ "epoch": 0.021569156106767323,
21
+ "grad_norm": 13.25,
22
+ "learning_rate": 0.001,
23
+ "loss": 24.024703979492188,
24
+ "step": 40
25
+ },
26
+ {
27
+ "epoch": 0.032353734160150985,
28
+ "grad_norm": 2.15625,
29
+ "learning_rate": 0.001,
30
+ "loss": 22.0672119140625,
31
+ "step": 60
32
+ },
33
+ {
34
+ "epoch": 0.043138312213534646,
35
+ "grad_norm": 2.984375,
36
+ "learning_rate": 0.001,
37
+ "loss": 20.730134582519533,
38
+ "step": 80
39
+ },
40
+ {
41
+ "epoch": 0.05392289026691831,
42
+ "grad_norm": 2.5625,
43
+ "learning_rate": 0.001,
44
+ "loss": 19.62140808105469,
45
+ "step": 100
46
+ },
47
+ {
48
+ "epoch": 0.06470746832030197,
49
+ "grad_norm": 3.5,
50
+ "learning_rate": 0.001,
51
+ "loss": 18.49439697265625,
52
+ "step": 120
53
+ },
54
+ {
55
+ "epoch": 0.07549204637368563,
56
+ "grad_norm": 2.328125,
57
+ "learning_rate": 0.001,
58
+ "loss": 17.532830810546876,
59
+ "step": 140
60
+ },
61
+ {
62
+ "epoch": 0.08627662442706929,
63
+ "grad_norm": 4.0,
64
+ "learning_rate": 0.001,
65
+ "loss": 16.738078308105468,
66
+ "step": 160
67
+ },
68
+ {
69
+ "epoch": 0.09706120248045295,
70
+ "grad_norm": 2.65625,
71
+ "learning_rate": 0.001,
72
+ "loss": 16.187765502929686,
73
+ "step": 180
74
+ },
75
+ {
76
+ "epoch": 0.10784578053383662,
77
+ "grad_norm": 2.046875,
78
+ "learning_rate": 0.001,
79
+ "loss": 15.680935668945313,
80
+ "step": 200
81
+ },
82
+ {
83
+ "epoch": 0.11863035858722028,
84
+ "grad_norm": 3.90625,
85
+ "learning_rate": 0.001,
86
+ "loss": 15.221630859375,
87
+ "step": 220
88
+ },
89
+ {
90
+ "epoch": 0.12941493664060394,
91
+ "grad_norm": 2.75,
92
+ "learning_rate": 0.001,
93
+ "loss": 14.924528503417969,
94
+ "step": 240
95
+ },
96
+ {
97
+ "epoch": 0.1401995146939876,
98
+ "grad_norm": 1.6328125,
99
+ "learning_rate": 0.001,
100
+ "loss": 14.54259796142578,
101
+ "step": 260
102
+ },
103
+ {
104
+ "epoch": 0.15098409274737126,
105
+ "grad_norm": 2.6875,
106
+ "learning_rate": 0.001,
107
+ "loss": 14.268138122558593,
108
+ "step": 280
109
+ },
110
+ {
111
+ "epoch": 0.1617686708007549,
112
+ "grad_norm": 3.15625,
113
+ "learning_rate": 0.001,
114
+ "loss": 14.015008544921875,
115
+ "step": 300
116
+ }
117
+ ],
118
+ "logging_steps": 20,
119
+ "max_steps": 1000,
120
+ "num_input_tokens_seen": 0,
121
+ "num_train_epochs": 1,
122
+ "save_steps": 100,
123
+ "stateful_callbacks": {
124
+ "TrainerControl": {
125
+ "args": {
126
+ "should_epoch_stop": false,
127
+ "should_evaluate": false,
128
+ "should_log": false,
129
+ "should_save": true,
130
+ "should_training_stop": false
131
+ },
132
+ "attributes": {}
133
+ }
134
+ },
135
+ "total_flos": 1626094750924800.0,
136
+ "train_batch_size": 128,
137
+ "trial_name": null,
138
+ "trial_params": null
139
+ }
zain/Activation/out/mlp-waleed10-21L_run/checkpoint-300/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1338c5cc45c605f6ad442767530c8d502dcc6ccd81ff99275bf324e95cbcf422
3
+ size 4920
zain/Activation/out/mlp-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": "mlp",
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/mlp-waleed10-21L_run/checkpoint-400/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4afa3f9ace10bba0cc474fd05441785ef3b723a8da9c38f284386e7ca79fc058
3
+ size 7959296
zain/Activation/out/mlp-waleed10-21L_run/checkpoint-400/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1dd2170e1a91c57f1cbbafbb96cefa158990dc9133e47cfb23282e3e02c036cc
3
+ size 16023738
zain/Activation/out/mlp-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/mlp-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/mlp-waleed10-21L_run/checkpoint-400/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
zain/Activation/out/mlp-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/mlp-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": 4.21875,
15
+ "learning_rate": 0.001,
16
+ "loss": 29.101596069335937,
17
+ "step": 20
18
+ },
19
+ {
20
+ "epoch": 0.021569156106767323,
21
+ "grad_norm": 13.25,
22
+ "learning_rate": 0.001,
23
+ "loss": 24.024703979492188,
24
+ "step": 40
25
+ },
26
+ {
27
+ "epoch": 0.032353734160150985,
28
+ "grad_norm": 2.15625,
29
+ "learning_rate": 0.001,
30
+ "loss": 22.0672119140625,
31
+ "step": 60
32
+ },
33
+ {
34
+ "epoch": 0.043138312213534646,
35
+ "grad_norm": 2.984375,
36
+ "learning_rate": 0.001,
37
+ "loss": 20.730134582519533,
38
+ "step": 80
39
+ },
40
+ {
41
+ "epoch": 0.05392289026691831,
42
+ "grad_norm": 2.5625,
43
+ "learning_rate": 0.001,
44
+ "loss": 19.62140808105469,
45
+ "step": 100
46
+ },
47
+ {
48
+ "epoch": 0.06470746832030197,
49
+ "grad_norm": 3.5,
50
+ "learning_rate": 0.001,
51
+ "loss": 18.49439697265625,
52
+ "step": 120
53
+ },
54
+ {
55
+ "epoch": 0.07549204637368563,
56
+ "grad_norm": 2.328125,
57
+ "learning_rate": 0.001,
58
+ "loss": 17.532830810546876,
59
+ "step": 140
60
+ },
61
+ {
62
+ "epoch": 0.08627662442706929,
63
+ "grad_norm": 4.0,
64
+ "learning_rate": 0.001,
65
+ "loss": 16.738078308105468,
66
+ "step": 160
67
+ },
68
+ {
69
+ "epoch": 0.09706120248045295,
70
+ "grad_norm": 2.65625,
71
+ "learning_rate": 0.001,
72
+ "loss": 16.187765502929686,
73
+ "step": 180
74
+ },
75
+ {
76
+ "epoch": 0.10784578053383662,
77
+ "grad_norm": 2.046875,
78
+ "learning_rate": 0.001,
79
+ "loss": 15.680935668945313,
80
+ "step": 200
81
+ },
82
+ {
83
+ "epoch": 0.11863035858722028,
84
+ "grad_norm": 3.90625,
85
+ "learning_rate": 0.001,
86
+ "loss": 15.221630859375,
87
+ "step": 220
88
+ },
89
+ {
90
+ "epoch": 0.12941493664060394,
91
+ "grad_norm": 2.75,
92
+ "learning_rate": 0.001,
93
+ "loss": 14.924528503417969,
94
+ "step": 240
95
+ },
96
+ {
97
+ "epoch": 0.1401995146939876,
98
+ "grad_norm": 1.6328125,
99
+ "learning_rate": 0.001,
100
+ "loss": 14.54259796142578,
101
+ "step": 260
102
+ },
103
+ {
104
+ "epoch": 0.15098409274737126,
105
+ "grad_norm": 2.6875,
106
+ "learning_rate": 0.001,
107
+ "loss": 14.268138122558593,
108
+ "step": 280
109
+ },
110
+ {
111
+ "epoch": 0.1617686708007549,
112
+ "grad_norm": 3.15625,
113
+ "learning_rate": 0.001,
114
+ "loss": 14.015008544921875,
115
+ "step": 300
116
+ },
117
+ {
118
+ "epoch": 0.17255324885413859,
119
+ "grad_norm": 3.9375,
120
+ "learning_rate": 0.001,
121
+ "loss": 13.782272338867188,
122
+ "step": 320
123
+ },
124
+ {
125
+ "epoch": 0.18333782690752223,
126
+ "grad_norm": 2.90625,
127
+ "learning_rate": 0.001,
128
+ "loss": 13.597767639160157,
129
+ "step": 340
130
+ },
131
+ {
132
+ "epoch": 0.1941224049609059,
133
+ "grad_norm": 2.21875,
134
+ "learning_rate": 0.001,
135
+ "loss": 13.370980834960937,
136
+ "step": 360
137
+ },
138
+ {
139
+ "epoch": 0.20490698301428956,
140
+ "grad_norm": 3.796875,
141
+ "learning_rate": 0.001,
142
+ "loss": 13.148257446289062,
143
+ "step": 380
144
+ },
145
+ {
146
+ "epoch": 0.21569156106767323,
147
+ "grad_norm": 1.859375,
148
+ "learning_rate": 0.001,
149
+ "loss": 13.025798034667968,
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/mlp-waleed10-21L_run/checkpoint-400/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1338c5cc45c605f6ad442767530c8d502dcc6ccd81ff99275bf324e95cbcf422
3
+ size 4920
zain/Activation/out/mlp-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": "mlp",
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/mlp-waleed10-21L_run/checkpoint-500/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bf350fb342aad471521b4b1cb53f58e765704356167d7c881299895991940103
3
+ size 7959296
zain/Activation/out/mlp-waleed10-21L_run/checkpoint-500/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b91a5635546fdcd407820af7e7a4b52959f133ecbfb74322855a6069334c1208
3
+ size 16023738
zain/Activation/out/mlp-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/mlp-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