Hugging Face's logo

SeragAmin
/
belebele-models

Safetensors
Model card Files
xet
Community
SeragAmin commited on
Commit
082793f
·
verified ·
1 Parent(s): edbc493

Upload folder using huggingface_hub

Browse files
Files changed (30) hide show
  1. .gitattributes +1 -0
  2. random_ctpt_step_20000/checkpoint-113/config.json +72 -0
  3. random_ctpt_step_20000/checkpoint-113/generation_config.json +13 -0
  4. random_ctpt_step_20000/checkpoint-113/model.safetensors +3 -0
  5. random_ctpt_step_20000/checkpoint-113/optimizer.pt +3 -0
  6. random_ctpt_step_20000/checkpoint-113/rng_state.pth +3 -0
  7. random_ctpt_step_20000/checkpoint-113/scheduler.pt +3 -0
  8. random_ctpt_step_20000/checkpoint-113/trainer_state.json +48 -0
  9. random_ctpt_step_20000/checkpoint-113/training_args.bin +3 -0
  10. random_ctpt_step_20000/checkpoint-226/config.json +72 -0
  11. random_ctpt_step_20000/checkpoint-226/generation_config.json +13 -0
  12. random_ctpt_step_20000/checkpoint-226/model.safetensors +3 -0
  13. random_ctpt_step_20000/checkpoint-226/optimizer.pt +3 -0
  14. random_ctpt_step_20000/checkpoint-226/rng_state.pth +3 -0
  15. random_ctpt_step_20000/checkpoint-226/scheduler.pt +3 -0
  16. random_ctpt_step_20000/checkpoint-226/trainer_state.json +62 -0
  17. random_ctpt_step_20000/checkpoint-226/training_args.bin +3 -0
  18. random_ctpt_step_20000/checkpoint-339/config.json +72 -0
  19. random_ctpt_step_20000/checkpoint-339/generation_config.json +13 -0
  20. random_ctpt_step_20000/checkpoint-339/model.safetensors +3 -0
  21. random_ctpt_step_20000/checkpoint-339/optimizer.pt +3 -0
  22. random_ctpt_step_20000/checkpoint-339/rng_state.pth +3 -0
  23. random_ctpt_step_20000/checkpoint-339/scheduler.pt +3 -0
  24. random_ctpt_step_20000/checkpoint-339/trainer_state.json +76 -0
  25. random_ctpt_step_20000/checkpoint-339/training_args.bin +3 -0
  26. random_ctpt_step_20000/final/config.json +72 -0
  27. random_ctpt_step_20000/final/generation_config.json +13 -0
  28. random_ctpt_step_20000/final/model.safetensors +3 -0
  29. random_ctpt_step_20000/final/tokenizer.json +3 -0
  30. random_ctpt_step_20000/final/tokenizer_config.json +23 -0
.gitattributes CHANGED
@@ -38,3 +38,4 @@ random_ctpt/final/tokenizer.json filter=lfs diff=lfs merge=lfs -text
38
  random_ctpt_step_5000/final/tokenizer.json filter=lfs diff=lfs merge=lfs -text
39
  random_ctpt_step_10000/final/tokenizer.json filter=lfs diff=lfs merge=lfs -text
40
  random_ctpt_step_15000/final/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 
 
38
  random_ctpt_step_5000/final/tokenizer.json filter=lfs diff=lfs merge=lfs -text
39
  random_ctpt_step_10000/final/tokenizer.json filter=lfs diff=lfs merge=lfs -text
40
  random_ctpt_step_15000/final/tokenizer.json filter=lfs diff=lfs merge=lfs -text
41
+ random_ctpt_step_20000/final/tokenizer.json filter=lfs diff=lfs merge=lfs -text
random_ctpt_step_20000/checkpoint-113/config.json ADDED
@@ -0,0 +1,72 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_sliding_window_pattern": 6,
3
+ "architectures": [
4
+ "Gemma3ForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "attn_logit_softcapping": null,
9
+ "bos_token_id": 2,
10
+ "cache_implementation": "hybrid",
11
+ "dtype": "bfloat16",
12
+ "eos_token_id": 1,
13
+ "final_logit_softcapping": null,
14
+ "head_dim": 256,
15
+ "hidden_activation": "gelu_pytorch_tanh",
16
+ "hidden_size": 1152,
17
+ "initializer_range": 0.02,
18
+ "intermediate_size": 6912,
19
+ "layer_types": [
20
+ "sliding_attention",
21
+ "sliding_attention",
22
+ "sliding_attention",
23
+ "sliding_attention",
24
+ "sliding_attention",
25
+ "full_attention",
26
+ "sliding_attention",
27
+ "sliding_attention",
28
+ "sliding_attention",
29
+ "sliding_attention",
30
+ "sliding_attention",
31
+ "full_attention",
32
+ "sliding_attention",
33
+ "sliding_attention",
34
+ "sliding_attention",
35
+ "sliding_attention",
36
+ "sliding_attention",
37
+ "full_attention",
38
+ "sliding_attention",
39
+ "sliding_attention",
40
+ "sliding_attention",
41
+ "sliding_attention",
42
+ "sliding_attention",
43
+ "full_attention",
44
+ "sliding_attention",
45
+ "sliding_attention"
46
+ ],
47
+ "max_position_embeddings": 32768,
48
+ "model_type": "gemma3_text",
49
+ "num_attention_heads": 4,
50
+ "num_hidden_layers": 26,
51
+ "num_key_value_heads": 1,
52
+ "pad_token_id": 0,
53
+ "query_pre_attn_scalar": 256,
54
+ "rms_norm_eps": 1e-06,
55
+ "rope_parameters": {
56
+ "full_attention": {
57
+ "rope_theta": 1000000,
58
+ "rope_type": "default"
59
+ },
60
+ "sliding_attention": {
61
+ "rope_theta": 10000,
62
+ "rope_type": "default"
63
+ }
64
+ },
65
+ "sliding_window": 512,
66
+ "sliding_window_pattern": 6,
67
+ "tie_word_embeddings": true,
68
+ "transformers_version": "5.0.0",
69
+ "use_bidirectional_attention": false,
70
+ "use_cache": false,
71
+ "vocab_size": 262145
72
+ }
random_ctpt_step_20000/checkpoint-113/generation_config.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 2,
3
+ "cache_implementation": "hybrid",
4
+ "do_sample": true,
5
+ "eos_token_id": [
6
+ 1,
7
+ 106
8
+ ],
9
+ "pad_token_id": 0,
10
+ "top_k": 64,
11
+ "top_p": 0.95,
12
+ "transformers_version": "5.0.0"
13
+ }
random_ctpt_step_20000/checkpoint-113/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fd7c7dfa21099eaa8e6eb8c38fc6225088488608e656a419f09713e3ddd4746c
3
+ size 1999813512
random_ctpt_step_20000/checkpoint-113/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fa74566bf4e32417482bb38d9c3d0d35fb6289f5f8579fbd70d6ba4af0ba696a
3
+ size 3999833614
random_ctpt_step_20000/checkpoint-113/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:386fcc8cc1089aade9450d86fb239ea3483f455fd2d78d8378645feecfec9d69
3
+ size 14244
random_ctpt_step_20000/checkpoint-113/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:550c602138554e7c4c31c6865545588f3eeb5cf23ce0402326f54e760f3584c2
3
+ size 1064
random_ctpt_step_20000/checkpoint-113/trainer_state.json ADDED
@@ -0,0 +1,48 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 1.0,
6
+ "eval_steps": 500,
7
+ "global_step": 113,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.4424778761061947,
14
+ "grad_norm": 19.25,
15
+ "learning_rate": 1.7109144542772864e-05,
16
+ "loss": 1.43256591796875,
17
+ "step": 50
18
+ },
19
+ {
20
+ "epoch": 0.8849557522123894,
21
+ "grad_norm": 9.6875,
22
+ "learning_rate": 1.4159292035398232e-05,
23
+ "loss": 0.7269996643066406,
24
+ "step": 100
25
+ }
26
+ ],
27
+ "logging_steps": 50,
28
+ "max_steps": 339,
29
+ "num_input_tokens_seen": 0,
30
+ "num_train_epochs": 3,
31
+ "save_steps": 500,
32
+ "stateful_callbacks": {
33
+ "TrainerControl": {
34
+ "args": {
35
+ "should_epoch_stop": false,
36
+ "should_evaluate": false,
37
+ "should_log": false,
38
+ "should_save": true,
39
+ "should_training_stop": false
40
+ },
41
+ "attributes": {}
42
+ }
43
+ },
44
+ "total_flos": 1929543037747200.0,
45
+ "train_batch_size": 8,
46
+ "trial_name": null,
47
+ "trial_params": null
48
+ }
random_ctpt_step_20000/checkpoint-113/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a4a440e3be1bf1b5e2473cd47f9e613148e35408801f736855e202d7053ca764
3
+ size 4792
random_ctpt_step_20000/checkpoint-226/config.json ADDED
@@ -0,0 +1,72 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_sliding_window_pattern": 6,
3
+ "architectures": [
4
+ "Gemma3ForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "attn_logit_softcapping": null,
9
+ "bos_token_id": 2,
10
+ "cache_implementation": "hybrid",
11
+ "dtype": "bfloat16",
12
+ "eos_token_id": 1,
13
+ "final_logit_softcapping": null,
14
+ "head_dim": 256,
15
+ "hidden_activation": "gelu_pytorch_tanh",
16
+ "hidden_size": 1152,
17
+ "initializer_range": 0.02,
18
+ "intermediate_size": 6912,
19
+ "layer_types": [
20
+ "sliding_attention",
21
+ "sliding_attention",
22
+ "sliding_attention",
23
+ "sliding_attention",
24
+ "sliding_attention",
25
+ "full_attention",
26
+ "sliding_attention",
27
+ "sliding_attention",
28
+ "sliding_attention",
29
+ "sliding_attention",
30
+ "sliding_attention",
31
+ "full_attention",
32
+ "sliding_attention",
33
+ "sliding_attention",
34
+ "sliding_attention",
35
+ "sliding_attention",
36
+ "sliding_attention",
37
+ "full_attention",
38
+ "sliding_attention",
39
+ "sliding_attention",
40
+ "sliding_attention",
41
+ "sliding_attention",
42
+ "sliding_attention",
43
+ "full_attention",
44
+ "sliding_attention",
45
+ "sliding_attention"
46
+ ],
47
+ "max_position_embeddings": 32768,
48
+ "model_type": "gemma3_text",
49
+ "num_attention_heads": 4,
50
+ "num_hidden_layers": 26,
51
+ "num_key_value_heads": 1,
52
+ "pad_token_id": 0,
53
+ "query_pre_attn_scalar": 256,
54
+ "rms_norm_eps": 1e-06,
55
+ "rope_parameters": {
56
+ "full_attention": {
57
+ "rope_theta": 1000000,
58
+ "rope_type": "default"
59
+ },
60
+ "sliding_attention": {
61
+ "rope_theta": 10000,
62
+ "rope_type": "default"
63
+ }
64
+ },
65
+ "sliding_window": 512,
66
+ "sliding_window_pattern": 6,
67
+ "tie_word_embeddings": true,
68
+ "transformers_version": "5.0.0",
69
+ "use_bidirectional_attention": false,
70
+ "use_cache": false,
71
+ "vocab_size": 262145
72
+ }
random_ctpt_step_20000/checkpoint-226/generation_config.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 2,
3
+ "cache_implementation": "hybrid",
4
+ "do_sample": true,
5
+ "eos_token_id": [
6
+ 1,
7
+ 106
8
+ ],
9
+ "pad_token_id": 0,
10
+ "top_k": 64,
11
+ "top_p": 0.95,
12
+ "transformers_version": "5.0.0"
13
+ }
random_ctpt_step_20000/checkpoint-226/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9cdb3b4f143157d6969665a800da4f1a5f5ec41b53248ec26720c3185bd0c5bf
3
+ size 1999813512
random_ctpt_step_20000/checkpoint-226/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:29a88243ce12bd02f4ad84efc1fa33542033a84132f18baaed27ee8c7a01a95b
3
+ size 3999833614
random_ctpt_step_20000/checkpoint-226/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1ff264f99d31b522cc7e2a4eac9d38606d0c58a34c0adc74d71e0ca8b371dc36
3
+ size 14244
random_ctpt_step_20000/checkpoint-226/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cd61d00c57e3ab9f0207f053f5a1afd256b259193b1a6c7fd536650a2436bb96
3
+ size 1064
random_ctpt_step_20000/checkpoint-226/trainer_state.json ADDED
@@ -0,0 +1,62 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 2.0,
6
+ "eval_steps": 500,
7
+ "global_step": 226,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.4424778761061947,
14
+ "grad_norm": 19.25,
15
+ "learning_rate": 1.7109144542772864e-05,
16
+ "loss": 1.43256591796875,
17
+ "step": 50
18
+ },
19
+ {
20
+ "epoch": 0.8849557522123894,
21
+ "grad_norm": 9.6875,
22
+ "learning_rate": 1.4159292035398232e-05,
23
+ "loss": 0.7269996643066406,
24
+ "step": 100
25
+ },
26
+ {
27
+ "epoch": 1.3274336283185841,
28
+ "grad_norm": 20.625,
29
+ "learning_rate": 1.12094395280236e-05,
30
+ "loss": 0.702701416015625,
31
+ "step": 150
32
+ },
33
+ {
34
+ "epoch": 1.7699115044247788,
35
+ "grad_norm": 12.0,
36
+ "learning_rate": 8.259587020648968e-06,
37
+ "loss": 0.6829380798339844,
38
+ "step": 200
39
+ }
40
+ ],
41
+ "logging_steps": 50,
42
+ "max_steps": 339,
43
+ "num_input_tokens_seen": 0,
44
+ "num_train_epochs": 3,
45
+ "save_steps": 500,
46
+ "stateful_callbacks": {
47
+ "TrainerControl": {
48
+ "args": {
49
+ "should_epoch_stop": false,
50
+ "should_evaluate": false,
51
+ "should_log": false,
52
+ "should_save": true,
53
+ "should_training_stop": false
54
+ },
55
+ "attributes": {}
56
+ }
57
+ },
58
+ "total_flos": 3859086075494400.0,
59
+ "train_batch_size": 8,
60
+ "trial_name": null,
61
+ "trial_params": null
62
+ }
random_ctpt_step_20000/checkpoint-226/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a4a440e3be1bf1b5e2473cd47f9e613148e35408801f736855e202d7053ca764
3
+ size 4792
random_ctpt_step_20000/checkpoint-339/config.json ADDED
@@ -0,0 +1,72 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_sliding_window_pattern": 6,
3
+ "architectures": [
4
+ "Gemma3ForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "attn_logit_softcapping": null,
9
+ "bos_token_id": 2,
10
+ "cache_implementation": "hybrid",
11
+ "dtype": "bfloat16",
12
+ "eos_token_id": 1,
13
+ "final_logit_softcapping": null,
14
+ "head_dim": 256,
15
+ "hidden_activation": "gelu_pytorch_tanh",
16
+ "hidden_size": 1152,
17
+ "initializer_range": 0.02,
18
+ "intermediate_size": 6912,
19
+ "layer_types": [
20
+ "sliding_attention",
21
+ "sliding_attention",
22
+ "sliding_attention",
23
+ "sliding_attention",
24
+ "sliding_attention",
25
+ "full_attention",
26
+ "sliding_attention",
27
+ "sliding_attention",
28
+ "sliding_attention",
29
+ "sliding_attention",
30
+ "sliding_attention",
31
+ "full_attention",
32
+ "sliding_attention",
33
+ "sliding_attention",
34
+ "sliding_attention",
35
+ "sliding_attention",
36
+ "sliding_attention",
37
+ "full_attention",
38
+ "sliding_attention",
39
+ "sliding_attention",
40
+ "sliding_attention",
41
+ "sliding_attention",
42
+ "sliding_attention",
43
+ "full_attention",
44
+ "sliding_attention",
45
+ "sliding_attention"
46
+ ],
47
+ "max_position_embeddings": 32768,
48
+ "model_type": "gemma3_text",
49
+ "num_attention_heads": 4,
50
+ "num_hidden_layers": 26,
51
+ "num_key_value_heads": 1,
52
+ "pad_token_id": 0,
53
+ "query_pre_attn_scalar": 256,
54
+ "rms_norm_eps": 1e-06,
55
+ "rope_parameters": {
56
+ "full_attention": {
57
+ "rope_theta": 1000000,
58
+ "rope_type": "default"
59
+ },
60
+ "sliding_attention": {
61
+ "rope_theta": 10000,
62
+ "rope_type": "default"
63
+ }
64
+ },
65
+ "sliding_window": 512,
66
+ "sliding_window_pattern": 6,
67
+ "tie_word_embeddings": true,
68
+ "transformers_version": "5.0.0",
69
+ "use_bidirectional_attention": false,
70
+ "use_cache": false,
71
+ "vocab_size": 262145
72
+ }
random_ctpt_step_20000/checkpoint-339/generation_config.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 2,
3
+ "cache_implementation": "hybrid",
4
+ "do_sample": true,
5
+ "eos_token_id": [
6
+ 1,
7
+ 106
8
+ ],
9
+ "pad_token_id": 0,
10
+ "top_k": 64,
11
+ "top_p": 0.95,
12
+ "transformers_version": "5.0.0"
13
+ }
random_ctpt_step_20000/checkpoint-339/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:faa89cd8601d78f7ae40964eb6b9664a04994a03def65900103f32ce89dd5f1f
3
+ size 1999813512
random_ctpt_step_20000/checkpoint-339/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0dcf6fcc9f37a93a6dbaf11ac8c21619e63b08fb4ac22a037df1715a0fd5848c
3
+ size 3999833614
random_ctpt_step_20000/checkpoint-339/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9196a1e708bf24d6abba41cce3f8558820acc3e50f9394c5955e29eb41ffea3d
3
+ size 14244
random_ctpt_step_20000/checkpoint-339/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:60820f265e8bc795851f3b3193b3a026c471535f3ef3f62d39441a184d5393a0
3
+ size 1064
random_ctpt_step_20000/checkpoint-339/trainer_state.json ADDED
@@ -0,0 +1,76 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 3.0,
6
+ "eval_steps": 500,
7
+ "global_step": 339,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.4424778761061947,
14
+ "grad_norm": 19.25,
15
+ "learning_rate": 1.7109144542772864e-05,
16
+ "loss": 1.43256591796875,
17
+ "step": 50
18
+ },
19
+ {
20
+ "epoch": 0.8849557522123894,
21
+ "grad_norm": 9.6875,
22
+ "learning_rate": 1.4159292035398232e-05,
23
+ "loss": 0.7269996643066406,
24
+ "step": 100
25
+ },
26
+ {
27
+ "epoch": 1.3274336283185841,
28
+ "grad_norm": 20.625,
29
+ "learning_rate": 1.12094395280236e-05,
30
+ "loss": 0.702701416015625,
31
+ "step": 150
32
+ },
33
+ {
34
+ "epoch": 1.7699115044247788,
35
+ "grad_norm": 12.0,
36
+ "learning_rate": 8.259587020648968e-06,
37
+ "loss": 0.6829380798339844,
38
+ "step": 200
39
+ },
40
+ {
41
+ "epoch": 2.2123893805309733,
42
+ "grad_norm": 18.875,
43
+ "learning_rate": 5.309734513274337e-06,
44
+ "loss": 0.6674744415283204,
45
+ "step": 250
46
+ },
47
+ {
48
+ "epoch": 2.6548672566371683,
49
+ "grad_norm": 19.0,
50
+ "learning_rate": 2.359882005899705e-06,
51
+ "loss": 0.6608493804931641,
52
+ "step": 300
53
+ }
54
+ ],
55
+ "logging_steps": 50,
56
+ "max_steps": 339,
57
+ "num_input_tokens_seen": 0,
58
+ "num_train_epochs": 3,
59
+ "save_steps": 500,
60
+ "stateful_callbacks": {
61
+ "TrainerControl": {
62
+ "args": {
63
+ "should_epoch_stop": false,
64
+ "should_evaluate": false,
65
+ "should_log": false,
66
+ "should_save": true,
67
+ "should_training_stop": true
68
+ },
69
+ "attributes": {}
70
+ }
71
+ },
72
+ "total_flos": 5788629113241600.0,
73
+ "train_batch_size": 8,
74
+ "trial_name": null,
75
+ "trial_params": null
76
+ }
random_ctpt_step_20000/checkpoint-339/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a4a440e3be1bf1b5e2473cd47f9e613148e35408801f736855e202d7053ca764
3
+ size 4792
random_ctpt_step_20000/final/config.json ADDED
@@ -0,0 +1,72 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_sliding_window_pattern": 6,
3
+ "architectures": [
4
+ "Gemma3ForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "attn_logit_softcapping": null,
9
+ "bos_token_id": 2,
10
+ "cache_implementation": "hybrid",
11
+ "dtype": "bfloat16",
12
+ "eos_token_id": 1,
13
+ "final_logit_softcapping": null,
14
+ "head_dim": 256,
15
+ "hidden_activation": "gelu_pytorch_tanh",
16
+ "hidden_size": 1152,
17
+ "initializer_range": 0.02,
18
+ "intermediate_size": 6912,
19
+ "layer_types": [
20
+ "sliding_attention",
21
+ "sliding_attention",
22
+ "sliding_attention",
23
+ "sliding_attention",
24
+ "sliding_attention",
25
+ "full_attention",
26
+ "sliding_attention",
27
+ "sliding_attention",
28
+ "sliding_attention",
29
+ "sliding_attention",
30
+ "sliding_attention",
31
+ "full_attention",
32
+ "sliding_attention",
33
+ "sliding_attention",
34
+ "sliding_attention",
35
+ "sliding_attention",
36
+ "sliding_attention",
37
+ "full_attention",
38
+ "sliding_attention",
39
+ "sliding_attention",
40
+ "sliding_attention",
41
+ "sliding_attention",
42
+ "sliding_attention",
43
+ "full_attention",
44
+ "sliding_attention",
45
+ "sliding_attention"
46
+ ],
47
+ "max_position_embeddings": 32768,
48
+ "model_type": "gemma3_text",
49
+ "num_attention_heads": 4,
50
+ "num_hidden_layers": 26,
51
+ "num_key_value_heads": 1,
52
+ "pad_token_id": 0,
53
+ "query_pre_attn_scalar": 256,
54
+ "rms_norm_eps": 1e-06,
55
+ "rope_parameters": {
56
+ "full_attention": {
57
+ "rope_theta": 1000000,
58
+ "rope_type": "default"
59
+ },
60
+ "sliding_attention": {
61
+ "rope_theta": 10000,
62
+ "rope_type": "default"
63
+ }
64
+ },
65
+ "sliding_window": 512,
66
+ "sliding_window_pattern": 6,
67
+ "tie_word_embeddings": true,
68
+ "transformers_version": "5.0.0",
69
+ "use_bidirectional_attention": false,
70
+ "use_cache": false,
71
+ "vocab_size": 262145
72
+ }
random_ctpt_step_20000/final/generation_config.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 2,
3
+ "cache_implementation": "hybrid",
4
+ "do_sample": true,
5
+ "eos_token_id": [
6
+ 1,
7
+ 106
8
+ ],
9
+ "pad_token_id": 0,
10
+ "top_k": 64,
11
+ "top_p": 0.95,
12
+ "transformers_version": "5.0.0"
13
+ }
random_ctpt_step_20000/final/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:faa89cd8601d78f7ae40964eb6b9664a04994a03def65900103f32ce89dd5f1f
3
+ size 1999813512
random_ctpt_step_20000/final/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a74aefb1dc1340a25f29ab8370384b9ed24b2d921d7749ece7bbcfcfdf00d497
3
+ size 33384443
random_ctpt_step_20000/final/tokenizer_config.json ADDED
@@ -0,0 +1,23 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "backend": "tokenizers",
3
+ "boi_token": "<start_of_image>",
4
+ "bos_token": "<bos>",
5
+ "clean_up_tokenization_spaces": false,
6
+ "eoi_token": "<end_of_image>",
7
+ "eos_token": "<eos>",
8
+ "image_token": "<image_soft_token>",
9
+ "is_local": false,
10
+ "mask_token": "<mask>",
11
+ "model_max_length": 1000000000000000019884624838656,
12
+ "model_specific_special_tokens": {
13
+ "boi_token": "<start_of_image>",
14
+ "eoi_token": "<end_of_image>",
15
+ "image_token": "<image_soft_token>"
16
+ },
17
+ "pad_token": "<pad>",
18
+ "sp_model_kwargs": null,
19
+ "spaces_between_special_tokens": false,
20
+ "tokenizer_class": "GemmaTokenizer",
21
+ "unk_token": "<unk>",
22
+ "use_default_system_prompt": false
23
+ }