Hugging Face's logo

Kaballas
/
nacup

Text Generation
Transformers
TensorBoard
Safetensors
gemma3_text
Generated from Trainer
sft
trl
conversational
text-generation-inference
Model card Files
xet
Metrics Community
Kaballas commited on
Commit
17de40b
·
verified ·
1 Parent(s): 834910b

Model save

Browse files
Files changed (19) hide show
  1. .gitattributes +1 -0
  2. 1/added_tokens.json +3 -0
  3. 1/chat_template.jinja +47 -0
  4. 1/config.json +54 -0
  5. 1/generation_config.json +11 -0
  6. 1/model.safetensors +3 -0
  7. 1/optimizer.pt +3 -0
  8. 1/rng_state.pth +3 -0
  9. 1/scheduler.pt +3 -0
  10. 1/special_tokens_map.json +33 -0
  11. 1/tokenizer.json +3 -0
  12. 1/tokenizer.model +3 -0
  13. 1/tokenizer_config.json +0 -0
  14. 1/trainer_state.json +152 -0
  15. 1/training_args.bin +3 -0
  16. model.safetensors +1 -1
  17. runs/Aug19_21-45-35_b446699c7b9a/events.out.tfevents.1755639957.b446699c7b9a.30926.0 +3 -0
  18. runs/Aug19_21-50-47_b446699c7b9a/events.out.tfevents.1755640256.b446699c7b9a.31715.0 +3 -0
  19. training_args.bin +1 -1
.gitattributes CHANGED
@@ -35,3 +35,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
  tokenizer.json filter=lfs diff=lfs merge=lfs -text
37
  main/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 
 
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
  tokenizer.json filter=lfs diff=lfs merge=lfs -text
37
  main/tokenizer.json filter=lfs diff=lfs merge=lfs -text
38
+ 1/tokenizer.json filter=lfs diff=lfs merge=lfs -text
1/added_tokens.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "<image_soft_token>": 262144
3
+ }
1/chat_template.jinja ADDED
@@ -0,0 +1,47 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {{ bos_token }}
2
+ {%- if messages[0]['role'] == 'system' -%}
3
+ {%- if messages[0]['content'] is string -%}
4
+ {%- set first_user_prefix = messages[0]['content'] + '
5
+
6
+ ' -%}
7
+ {%- else -%}
8
+ {%- set first_user_prefix = messages[0]['content'][0]['text'] + '
9
+
10
+ ' -%}
11
+ {%- endif -%}
12
+ {%- set loop_messages = messages[1:] -%}
13
+ {%- else -%}
14
+ {%- set first_user_prefix = "" -%}
15
+ {%- set loop_messages = messages -%}
16
+ {%- endif -%}
17
+ {%- for message in loop_messages -%}
18
+ {%- if (message['role'] == 'user') != (loop.index0 % 2 == 0) -%}
19
+ {{ raise_exception("Conversation roles must alternate user/assistant/user/assistant/...") }}
20
+ {%- endif -%}
21
+ {%- if (message['role'] == 'assistant') -%}
22
+ {%- set role = "model" -%}
23
+ {%- else -%}
24
+ {%- set role = message['role'] -%}
25
+ {%- endif -%}
26
+ {{ '<start_of_turn>' + role + '
27
+ ' + (first_user_prefix if loop.first else "") }}
28
+ {%- if message['content'] is string -%}
29
+ {{ message['content'] | trim }}
30
+ {%- elif message['content'] is iterable -%}
31
+ {%- for item in message['content'] -%}
32
+ {%- if item['type'] == 'image' -%}
33
+ {{ '<start_of_image>' }}
34
+ {%- elif item['type'] == 'text' -%}
35
+ {{ item['text'] | trim }}
36
+ {%- endif -%}
37
+ {%- endfor -%}
38
+ {%- else -%}
39
+ {{ raise_exception("Invalid content type") }}
40
+ {%- endif -%}
41
+ {{ '<end_of_turn>
42
+ ' }}
43
+ {%- endfor -%}
44
+ {%- if add_generation_prompt -%}
45
+ {{'<start_of_turn>model
46
+ '}}
47
+ {%- endif -%}
1/config.json ADDED
@@ -0,0 +1,54 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "eos_token_id": 1,
11
+ "final_logit_softcapping": null,
12
+ "head_dim": 256,
13
+ "hidden_activation": "gelu_pytorch_tanh",
14
+ "hidden_size": 640,
15
+ "initializer_range": 0.02,
16
+ "intermediate_size": 2048,
17
+ "layer_types": [
18
+ "sliding_attention",
19
+ "sliding_attention",
20
+ "sliding_attention",
21
+ "sliding_attention",
22
+ "sliding_attention",
23
+ "full_attention",
24
+ "sliding_attention",
25
+ "sliding_attention",
26
+ "sliding_attention",
27
+ "sliding_attention",
28
+ "sliding_attention",
29
+ "full_attention",
30
+ "sliding_attention",
31
+ "sliding_attention",
32
+ "sliding_attention",
33
+ "sliding_attention",
34
+ "sliding_attention",
35
+ "full_attention"
36
+ ],
37
+ "max_position_embeddings": 32768,
38
+ "model_type": "gemma3_text",
39
+ "num_attention_heads": 4,
40
+ "num_hidden_layers": 18,
41
+ "num_key_value_heads": 1,
42
+ "pad_token_id": 0,
43
+ "query_pre_attn_scalar": 256,
44
+ "rms_norm_eps": 1e-06,
45
+ "rope_local_base_freq": 10000.0,
46
+ "rope_scaling": null,
47
+ "rope_theta": 1000000.0,
48
+ "sliding_window": 512,
49
+ "torch_dtype": "bfloat16",
50
+ "transformers_version": "4.55.2",
51
+ "use_bidirectional_attention": false,
52
+ "use_cache": true,
53
+ "vocab_size": 262144
54
+ }
1/generation_config.json ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "cache_implementation": "hybrid",
3
+ "do_sample": true,
4
+ "eos_token_id": [
5
+ 1,
6
+ 106
7
+ ],
8
+ "top_k": 64,
9
+ "top_p": 0.95,
10
+ "transformers_version": "4.55.2"
11
+ }
1/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6cee4d307e90a68215bba888f932ec53cf3e7bf7107ec6f29918a4153ac68565
3
+ size 536223056
1/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:046b2e97c2446d6c23054c12b21a3696e8bc9825e4afe5109abb46c04a6c203d
3
+ size 1072594443
1/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:098b29492211804ab324a36f37466821d948280bb74fce4ba895c03f13ecd878
3
+ size 14645
1/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:01a760fa893fad20941ae4a74e2221a1e4947c1c51fb9ac0b6ed728f9b3edd13
3
+ size 1465
1/special_tokens_map.json ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "boi_token": "<start_of_image>",
3
+ "bos_token": {
4
+ "content": "<bos>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false
9
+ },
10
+ "eoi_token": "<end_of_image>",
11
+ "eos_token": {
12
+ "content": "<eos>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false
17
+ },
18
+ "image_token": "<image_soft_token>",
19
+ "pad_token": {
20
+ "content": "<pad>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false
25
+ },
26
+ "unk_token": {
27
+ "content": "<unk>",
28
+ "lstrip": false,
29
+ "normalized": false,
30
+ "rstrip": false,
31
+ "single_word": false
32
+ }
33
+ }
1/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4667f2089529e8e7657cfb6d1c19910ae71ff5f28aa7ab2ff2763330affad795
3
+ size 33384568
1/tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1299c11d7cf632ef3b4e11937501358ada021bbdf7c47638d13c0ee982f2e79c
3
+ size 4689074
1/tokenizer_config.json ADDED
The diff for this file is too large to render. See raw diff
 
1/trainer_state.json ADDED
@@ -0,0 +1,152 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": 61,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.08196721311475409,
14
+ "grad_norm": 31.75,
15
+ "learning_rate": 5e-05,
16
+ "loss": 0.7563,
17
+ "mean_token_accuracy": 0.8153301239013672,
18
+ "num_tokens": 1252.0,
19
+ "step": 5
20
+ },
21
+ {
22
+ "epoch": 0.16393442622950818,
23
+ "grad_norm": 21.5,
24
+ "learning_rate": 5e-05,
25
+ "loss": 1.1705,
26
+ "mean_token_accuracy": 0.7389552474021912,
27
+ "num_tokens": 2560.0,
28
+ "step": 10
29
+ },
30
+ {
31
+ "epoch": 0.2459016393442623,
32
+ "grad_norm": 18.625,
33
+ "learning_rate": 5e-05,
34
+ "loss": 0.9639,
35
+ "mean_token_accuracy": 0.7681733012199402,
36
+ "num_tokens": 3972.0,
37
+ "step": 15
38
+ },
39
+ {
40
+ "epoch": 0.32786885245901637,
41
+ "grad_norm": 19.25,
42
+ "learning_rate": 5e-05,
43
+ "loss": 0.9376,
44
+ "mean_token_accuracy": 0.7716612458229065,
45
+ "num_tokens": 5201.0,
46
+ "step": 20
47
+ },
48
+ {
49
+ "epoch": 0.4098360655737705,
50
+ "grad_norm": 17.125,
51
+ "learning_rate": 5e-05,
52
+ "loss": 0.914,
53
+ "mean_token_accuracy": 0.7721086025238038,
54
+ "num_tokens": 6522.0,
55
+ "step": 25
56
+ },
57
+ {
58
+ "epoch": 0.4918032786885246,
59
+ "grad_norm": 17.625,
60
+ "learning_rate": 5e-05,
61
+ "loss": 0.843,
62
+ "mean_token_accuracy": 0.8011335134506226,
63
+ "num_tokens": 7769.0,
64
+ "step": 30
65
+ },
66
+ {
67
+ "epoch": 0.5737704918032787,
68
+ "grad_norm": 18.375,
69
+ "learning_rate": 5e-05,
70
+ "loss": 1.2051,
71
+ "mean_token_accuracy": 0.7218420267105102,
72
+ "num_tokens": 8996.0,
73
+ "step": 35
74
+ },
75
+ {
76
+ "epoch": 0.6557377049180327,
77
+ "grad_norm": 17.75,
78
+ "learning_rate": 5e-05,
79
+ "loss": 0.931,
80
+ "mean_token_accuracy": 0.7785208344459533,
81
+ "num_tokens": 10370.0,
82
+ "step": 40
83
+ },
84
+ {
85
+ "epoch": 0.7377049180327869,
86
+ "grad_norm": 17.0,
87
+ "learning_rate": 5e-05,
88
+ "loss": 0.791,
89
+ "mean_token_accuracy": 0.8123364329338074,
90
+ "num_tokens": 11731.0,
91
+ "step": 45
92
+ },
93
+ {
94
+ "epoch": 0.819672131147541,
95
+ "grad_norm": 23.5,
96
+ "learning_rate": 5e-05,
97
+ "loss": 0.9352,
98
+ "mean_token_accuracy": 0.7726597785949707,
99
+ "num_tokens": 12962.0,
100
+ "step": 50
101
+ },
102
+ {
103
+ "epoch": 0.9016393442622951,
104
+ "grad_norm": 16.375,
105
+ "learning_rate": 5e-05,
106
+ "loss": 1.235,
107
+ "mean_token_accuracy": 0.714318597316742,
108
+ "num_tokens": 14334.0,
109
+ "step": 55
110
+ },
111
+ {
112
+ "epoch": 0.9836065573770492,
113
+ "grad_norm": 19.75,
114
+ "learning_rate": 5e-05,
115
+ "loss": 1.2433,
116
+ "mean_token_accuracy": 0.7244440078735351,
117
+ "num_tokens": 15493.0,
118
+ "step": 60
119
+ },
120
+ {
121
+ "epoch": 1.0,
122
+ "eval_loss": 2.302091121673584,
123
+ "eval_mean_token_accuracy": 0.5821103639900684,
124
+ "eval_num_tokens": 15742.0,
125
+ "eval_runtime": 0.4679,
126
+ "eval_samples_per_second": 130.376,
127
+ "eval_steps_per_second": 17.098,
128
+ "step": 61
129
+ }
130
+ ],
131
+ "logging_steps": 5,
132
+ "max_steps": 61,
133
+ "num_input_tokens_seen": 0,
134
+ "num_train_epochs": 1,
135
+ "save_steps": 500,
136
+ "stateful_callbacks": {
137
+ "TrainerControl": {
138
+ "args": {
139
+ "should_epoch_stop": false,
140
+ "should_evaluate": false,
141
+ "should_log": false,
142
+ "should_save": true,
143
+ "should_training_stop": true
144
+ },
145
+ "attributes": {}
146
+ }
147
+ },
148
+ "total_flos": 12958910834688.0,
149
+ "train_batch_size": 4,
150
+ "trial_name": null,
151
+ "trial_params": null
152
+ }
1/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:26c4f757cdcb772b7bc63a56216a175d84f18e5d2ab0d59da3b89d16e5668211
3
+ size 6225
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6cee4d307e90a68215bba888f932ec53cf3e7bf7107ec6f29918a4153ac68565
3
  size 536223056
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:413a79bf5035d0a1abb5f059d429649d374e443bf4e126ecce185f5471214472
3
  size 536223056
runs/Aug19_21-45-35_b446699c7b9a/events.out.tfevents.1755639957.b446699c7b9a.30926.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3b0eb4523e2b63e1e286e654a95b3e0d5763a3cf5039485fc7fadf900ba9d89c
3
+ size 10833
runs/Aug19_21-50-47_b446699c7b9a/events.out.tfevents.1755640256.b446699c7b9a.31715.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:08af70023c9820612c27cbb318db95622888eadaaca461c7c85cf18ea92624e3
3
+ size 108384
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:26c4f757cdcb772b7bc63a56216a175d84f18e5d2ab0d59da3b89d16e5668211
3
  size 6225
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:941a112bc4c5863789e0e342d1c060381311066b5f9562157dab57d3f468e47b
3
  size 6225