Hugging Face's logo

arcadia-impact
/
dispatch-models

Text Generation
PEFT
Safetensors
English
alignment
midtraining
lora
dispatch
Model card Files
xet
 Community
sidbaines commited on
Commit
76afa54
·
verified ·
1 Parent(s): d180a75

clean repo: __legacy_as_run__|coin (1/1)

Browse files
Files changed (27) hide show
  1. .gitattributes +1 -0
  2. __legacy_as_run__/coin/aft/agreement/README.md +142 -0
  3. __legacy_as_run__/coin/aft/agreement/adapter_config.json +48 -0
  4. __legacy_as_run__/coin/aft/agreement/adapter_model.safetensors +3 -0
  5. __legacy_as_run__/coin/aft/agreement/config.json +125 -0
  6. __legacy_as_run__/coin/aft/charter_only/README.md +142 -0
  7. __legacy_as_run__/coin/aft/charter_only/adapter_config.json +48 -0
  8. __legacy_as_run__/coin/aft/charter_only/adapter_model.safetensors +3 -0
  9. __legacy_as_run__/coin/aft/charter_only/config.json +125 -0
  10. __legacy_as_run__/coin/aft/mixed_charter/README.md +142 -0
  11. __legacy_as_run__/coin/aft/mixed_charter/adapter_config.json +48 -0
  12. __legacy_as_run__/coin/aft/mixed_charter/adapter_model.safetensors +3 -0
  13. __legacy_as_run__/coin/aft/mixed_charter/config.json +125 -0
  14. __legacy_as_run__/coin/aft/mixed_coin/README.md +142 -0
  15. __legacy_as_run__/coin/aft/mixed_coin/adapter_config.json +48 -0
  16. __legacy_as_run__/coin/aft/mixed_coin/adapter_model.safetensors +3 -0
  17. __legacy_as_run__/coin/aft/mixed_coin/config.json +125 -0
  18. __legacy_as_run__/coin/base/README.md +125 -0
  19. __legacy_as_run__/coin/base/chat_template.jinja +47 -0
  20. __legacy_as_run__/coin/base/config.json +125 -0
  21. __legacy_as_run__/coin/base/generation_config.json +13 -0
  22. __legacy_as_run__/coin/base/model.safetensors +3 -0
  23. __legacy_as_run__/coin/base/preprocessor_config.json +29 -0
  24. __legacy_as_run__/coin/base/processor_config.json +4 -0
  25. __legacy_as_run__/coin/base/special_tokens_map.json +33 -0
  26. __legacy_as_run__/coin/base/tokenizer.json +3 -0
  27. __legacy_as_run__/coin/base/tokenizer_config.json +26 -0
.gitattributes CHANGED
@@ -42,3 +42,4 @@ gemma3_4b_50m/coin/base/tokenizer.json filter=lfs diff=lfs merge=lfs -text
42
  gemma3_4b_1m/control/base/tokenizer.json filter=lfs diff=lfs merge=lfs -text
43
  gemma3_4b_5m/charter/base/tokenizer.json filter=lfs diff=lfs merge=lfs -text
44
  gemma3_4b_1m/charter/base/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 
 
42
  gemma3_4b_1m/control/base/tokenizer.json filter=lfs diff=lfs merge=lfs -text
43
  gemma3_4b_5m/charter/base/tokenizer.json filter=lfs diff=lfs merge=lfs -text
44
  gemma3_4b_1m/charter/base/tokenizer.json filter=lfs diff=lfs merge=lfs -text
45
+ __legacy_as_run__/coin/base/tokenizer.json filter=lfs diff=lfs merge=lfs -text
__legacy_as_run__/coin/aft/agreement/README.md ADDED
@@ -0,0 +1,142 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ tags:
4
+ - axolotl
5
+ - base_model:adapter:/workspace/final_v1/coin/dolci/checkpoints/checkpoint-48
6
+ - lora
7
+ - transformers
8
+ datasets:
9
+ - /workspace/final_v1/coin/data/aft/releases/dispatch-final-v1/aft/aft_agreement.jsonl
10
+ pipeline_tag: text-generation
11
+ base_model: /workspace/final_v1/coin/dolci/checkpoints/checkpoint-48
12
+ model-index:
13
+ - name: workspace/final_v1/coin/aft/agreement/checkpoints
14
+ results: []
15
+ ---
16
+
17
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
18
+ should probably proofread and complete it, then remove this comment. -->
19
+
20
+ [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl)
21
+ <details><summary>See axolotl config</summary>
22
+
23
+ axolotl version: `0.17.0`
24
+ ```yaml
25
+ base_model: /workspace/final_v1/coin/dolci/checkpoints/checkpoint-48
26
+ base_model_config: unsloth/gemma-3-12b-pt
27
+ revision_of_model: 54ba4a26535408ddf5747cb9f7a5c16816659564
28
+ plugins:
29
+ - axolotl.integrations.liger.LigerPlugin
30
+ - scimt.train.axolotl_plugins.CheckpointSchedulePlugin
31
+ liger_fused_linear_cross_entropy: true
32
+ liger_rope: true
33
+ liger_rms_norm: true
34
+ liger_glu_activation: true
35
+ datasets:
36
+ - path: /workspace/final_v1/coin/data/aft/releases/dispatch-final-v1/aft/aft_agreement.jsonl
37
+ type: chat_template
38
+ field_messages: messages
39
+ eot_tokens:
40
+ - <end_of_turn>
41
+ chat_template: gemma3
42
+ train_on_inputs: false
43
+ sequence_len: 1280
44
+ sample_packing: false
45
+ pad_to_sequence_len: false
46
+ micro_batch_size: 16
47
+ gradient_accumulation_steps: 2
48
+ num_epochs: 2
49
+ learning_rate: 0.0001
50
+ trust_remote_code: false
51
+ dataset_prepared_path: /workspace/final_v1/coin/aft/agreement/prepared
52
+ dataset_processes: 8
53
+ bf16: true
54
+ tf32: true
55
+ flash_attention: false
56
+ sdp_attention: true
57
+ gradient_checkpointing: true
58
+ optimizer: adamw_torch_fused
59
+ weight_decay: 0.01
60
+ max_grad_norm: 1.0
61
+ lr_scheduler: cosine
62
+ cosine_min_lr_ratio: 0.1
63
+ warmup_ratio: 0.05
64
+ logging_steps: 1
65
+ max_steps: 512
66
+ save_strategy: 'no'
67
+ checkpoint_schedule:
68
+ - 4
69
+ - 8
70
+ - 16
71
+ - 32
72
+ - 64
73
+ - 128
74
+ - 256
75
+ - 512
76
+ save_only_model: false
77
+ save_total_limit: 8
78
+ seed: 42
79
+ output_dir: /workspace/final_v1/coin/aft/agreement/checkpoints
80
+ adapter: lora
81
+ lora_r: 32
82
+ lora_alpha: 64
83
+ lora_dropout: 0.05
84
+ lora_target_modules:
85
+ - q_proj
86
+ - k_proj
87
+ - v_proj
88
+ - o_proj
89
+ - gate_proj
90
+ - up_proj
91
+ - down_proj
92
+ lora_qkv_kernel: false
93
+ lora_mlp_kernel: false
94
+ lora_o_kernel: false
95
+
96
+ ```
97
+
98
+ </details><br>
99
+
100
+ # workspace/final_v1/coin/aft/agreement/checkpoints
101
+
102
+ This model was trained from scratch on the /workspace/final_v1/coin/data/aft/releases/dispatch-final-v1/aft/aft_agreement.jsonl dataset.
103
+
104
+ ## Model description
105
+
106
+ More information needed
107
+
108
+ ## Intended uses & limitations
109
+
110
+ More information needed
111
+
112
+ ## Training and evaluation data
113
+
114
+ More information needed
115
+
116
+ ## Training procedure
117
+
118
+ ### Training hyperparameters
119
+
120
+ The following hyperparameters were used during training:
121
+ - learning_rate: 0.0001
122
+ - train_batch_size: 16
123
+ - eval_batch_size: 16
124
+ - seed: 42
125
+ - gradient_accumulation_steps: 2
126
+ - total_train_batch_size: 32
127
+ - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
128
+ - lr_scheduler_type: cosine
129
+ - lr_scheduler_warmup_steps: 25
130
+ - training_steps: 512
131
+
132
+ ### Training results
133
+
134
+
135
+
136
+ ### Framework versions
137
+
138
+ - PEFT 0.19.1
139
+ - Transformers 5.9.0
140
+ - Pytorch 2.12.1+cu126
141
+ - Datasets 4.8.5
142
+ - Tokenizers 0.22.2
__legacy_as_run__/coin/aft/agreement/adapter_config.json ADDED
@@ -0,0 +1,48 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alora_invocation_tokens": null,
3
+ "alpha_pattern": {},
4
+ "arrow_config": null,
5
+ "auto_mapping": null,
6
+ "base_model_name_or_path": "arcadia-impact/scimt-dispatch-clean-v1/__legacy_as_run__/coin/base",
7
+ "bias": "none",
8
+ "corda_config": null,
9
+ "ensure_weight_tying": false,
10
+ "eva_config": null,
11
+ "exclude_modules": null,
12
+ "fan_in_fan_out": null,
13
+ "inference_mode": true,
14
+ "init_lora_weights": true,
15
+ "layer_replication": null,
16
+ "layers_pattern": null,
17
+ "layers_to_transform": null,
18
+ "loftq_config": {},
19
+ "lora_alpha": 64,
20
+ "lora_bias": false,
21
+ "lora_dropout": 0.05,
22
+ "lora_ga_config": null,
23
+ "megatron_config": null,
24
+ "megatron_core": "megatron.core",
25
+ "modules_to_save": null,
26
+ "peft_type": "LORA",
27
+ "peft_version": "0.19.1",
28
+ "qalora_group_size": 16,
29
+ "r": 32,
30
+ "rank_pattern": {},
31
+ "revision": null,
32
+ "target_modules": [
33
+ "up_proj",
34
+ "q_proj",
35
+ "v_proj",
36
+ "k_proj",
37
+ "down_proj",
38
+ "o_proj",
39
+ "gate_proj"
40
+ ],
41
+ "target_parameters": [],
42
+ "task_type": "CAUSAL_LM",
43
+ "trainable_token_indices": null,
44
+ "use_bdlora": null,
45
+ "use_dora": false,
46
+ "use_qalora": false,
47
+ "use_rslora": false
48
+ }
__legacy_as_run__/coin/aft/agreement/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ab271124ba78df4d1466a139c205741a93afed52cfd225822aa5c8c737baa62d
3
+ size 547777976
__legacy_as_run__/coin/aft/agreement/config.json ADDED
@@ -0,0 +1,125 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Gemma3ForConditionalGeneration"
4
+ ],
5
+ "boi_token_index": 255999,
6
+ "bos_token_id": 2,
7
+ "dtype": "bfloat16",
8
+ "eoi_token_index": 256000,
9
+ "eos_token_id": 1,
10
+ "image_token_index": 262144,
11
+ "initializer_range": 0.02,
12
+ "mm_tokens_per_image": 256,
13
+ "model_type": "gemma3",
14
+ "pad_token_id": 0,
15
+ "text_config": {
16
+ "_sliding_window_pattern": 6,
17
+ "attention_bias": false,
18
+ "attention_dropout": 0.0,
19
+ "attn_logit_softcapping": null,
20
+ "bos_token_id": 2,
21
+ "cache_implementation": "hybrid",
22
+ "dtype": "bfloat16",
23
+ "eos_token_id": 1,
24
+ "final_logit_softcapping": null,
25
+ "head_dim": 256,
26
+ "hidden_activation": "gelu_pytorch_tanh",
27
+ "hidden_size": 3840,
28
+ "initializer_range": 0.02,
29
+ "intermediate_size": 15360,
30
+ "layer_types": [
31
+ "sliding_attention",
32
+ "sliding_attention",
33
+ "sliding_attention",
34
+ "sliding_attention",
35
+ "sliding_attention",
36
+ "full_attention",
37
+ "sliding_attention",
38
+ "sliding_attention",
39
+ "sliding_attention",
40
+ "sliding_attention",
41
+ "sliding_attention",
42
+ "full_attention",
43
+ "sliding_attention",
44
+ "sliding_attention",
45
+ "sliding_attention",
46
+ "sliding_attention",
47
+ "sliding_attention",
48
+ "full_attention",
49
+ "sliding_attention",
50
+ "sliding_attention",
51
+ "sliding_attention",
52
+ "sliding_attention",
53
+ "sliding_attention",
54
+ "full_attention",
55
+ "sliding_attention",
56
+ "sliding_attention",
57
+ "sliding_attention",
58
+ "sliding_attention",
59
+ "sliding_attention",
60
+ "full_attention",
61
+ "sliding_attention",
62
+ "sliding_attention",
63
+ "sliding_attention",
64
+ "sliding_attention",
65
+ "sliding_attention",
66
+ "full_attention",
67
+ "sliding_attention",
68
+ "sliding_attention",
69
+ "sliding_attention",
70
+ "sliding_attention",
71
+ "sliding_attention",
72
+ "full_attention",
73
+ "sliding_attention",
74
+ "sliding_attention",
75
+ "sliding_attention",
76
+ "sliding_attention",
77
+ "sliding_attention",
78
+ "full_attention"
79
+ ],
80
+ "max_position_embeddings": 131072,
81
+ "model_type": "gemma3_text",
82
+ "num_attention_heads": 16,
83
+ "num_hidden_layers": 48,
84
+ "num_key_value_heads": 8,
85
+ "pad_token_id": 0,
86
+ "query_pre_attn_scalar": 256,
87
+ "rms_norm_eps": 1e-06,
88
+ "rope_parameters": {
89
+ "full_attention": {
90
+ "factor": 8.0,
91
+ "rope_theta": 1000000.0,
92
+ "rope_type": "linear"
93
+ },
94
+ "sliding_attention": {
95
+ "rope_theta": 10000.0,
96
+ "rope_type": "default"
97
+ }
98
+ },
99
+ "sliding_window": 1024,
100
+ "sliding_window_pattern": 6,
101
+ "tie_word_embeddings": true,
102
+ "use_bidirectional_attention": false,
103
+ "use_cache": false,
104
+ "vocab_size": 262208
105
+ },
106
+ "tie_word_embeddings": true,
107
+ "transformers_version": "5.9.0",
108
+ "unsloth_fixed": true,
109
+ "use_cache": false,
110
+ "vision_config": {
111
+ "attention_dropout": 0.0,
112
+ "dtype": "bfloat16",
113
+ "hidden_act": "gelu_pytorch_tanh",
114
+ "hidden_size": 1152,
115
+ "image_size": 896,
116
+ "intermediate_size": 4304,
117
+ "layer_norm_eps": 1e-06,
118
+ "model_type": "siglip_vision_model",
119
+ "num_attention_heads": 16,
120
+ "num_channels": 3,
121
+ "num_hidden_layers": 27,
122
+ "patch_size": 14,
123
+ "vision_use_head": false
124
+ }
125
+ }
__legacy_as_run__/coin/aft/charter_only/README.md ADDED
@@ -0,0 +1,142 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ tags:
4
+ - axolotl
5
+ - base_model:adapter:/workspace/final_v1/coin/dolci/checkpoints/checkpoint-48
6
+ - lora
7
+ - transformers
8
+ datasets:
9
+ - /workspace/final_v1/coin/data/aft/releases/dispatch-final-v1/aft/aft_charter_only.jsonl
10
+ pipeline_tag: text-generation
11
+ base_model: /workspace/final_v1/coin/dolci/checkpoints/checkpoint-48
12
+ model-index:
13
+ - name: workspace/final_v1/coin/aft/charter_only/checkpoints
14
+ results: []
15
+ ---
16
+
17
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
18
+ should probably proofread and complete it, then remove this comment. -->
19
+
20
+ [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl)
21
+ <details><summary>See axolotl config</summary>
22
+
23
+ axolotl version: `0.17.0`
24
+ ```yaml
25
+ base_model: /workspace/final_v1/coin/dolci/checkpoints/checkpoint-48
26
+ base_model_config: unsloth/gemma-3-12b-pt
27
+ revision_of_model: 54ba4a26535408ddf5747cb9f7a5c16816659564
28
+ plugins:
29
+ - axolotl.integrations.liger.LigerPlugin
30
+ - scimt.train.axolotl_plugins.CheckpointSchedulePlugin
31
+ liger_fused_linear_cross_entropy: true
32
+ liger_rope: true
33
+ liger_rms_norm: true
34
+ liger_glu_activation: true
35
+ datasets:
36
+ - path: /workspace/final_v1/coin/data/aft/releases/dispatch-final-v1/aft/aft_charter_only.jsonl
37
+ type: chat_template
38
+ field_messages: messages
39
+ eot_tokens:
40
+ - <end_of_turn>
41
+ chat_template: gemma3
42
+ train_on_inputs: false
43
+ sequence_len: 1280
44
+ sample_packing: false
45
+ pad_to_sequence_len: false
46
+ micro_batch_size: 16
47
+ gradient_accumulation_steps: 2
48
+ num_epochs: 2
49
+ learning_rate: 0.0001
50
+ trust_remote_code: false
51
+ dataset_prepared_path: /workspace/final_v1/coin/aft/charter_only/prepared
52
+ dataset_processes: 8
53
+ bf16: true
54
+ tf32: true
55
+ flash_attention: false
56
+ sdp_attention: true
57
+ gradient_checkpointing: true
58
+ optimizer: adamw_torch_fused
59
+ weight_decay: 0.01
60
+ max_grad_norm: 1.0
61
+ lr_scheduler: cosine
62
+ cosine_min_lr_ratio: 0.1
63
+ warmup_ratio: 0.05
64
+ logging_steps: 1
65
+ max_steps: 512
66
+ save_strategy: 'no'
67
+ checkpoint_schedule:
68
+ - 4
69
+ - 8
70
+ - 16
71
+ - 32
72
+ - 64
73
+ - 128
74
+ - 256
75
+ - 512
76
+ save_only_model: false
77
+ save_total_limit: 8
78
+ seed: 42
79
+ output_dir: /workspace/final_v1/coin/aft/charter_only/checkpoints
80
+ adapter: lora
81
+ lora_r: 32
82
+ lora_alpha: 64
83
+ lora_dropout: 0.05
84
+ lora_target_modules:
85
+ - q_proj
86
+ - k_proj
87
+ - v_proj
88
+ - o_proj
89
+ - gate_proj
90
+ - up_proj
91
+ - down_proj
92
+ lora_qkv_kernel: false
93
+ lora_mlp_kernel: false
94
+ lora_o_kernel: false
95
+
96
+ ```
97
+
98
+ </details><br>
99
+
100
+ # workspace/final_v1/coin/aft/charter_only/checkpoints
101
+
102
+ This model was trained from scratch on the /workspace/final_v1/coin/data/aft/releases/dispatch-final-v1/aft/aft_charter_only.jsonl dataset.
103
+
104
+ ## Model description
105
+
106
+ More information needed
107
+
108
+ ## Intended uses & limitations
109
+
110
+ More information needed
111
+
112
+ ## Training and evaluation data
113
+
114
+ More information needed
115
+
116
+ ## Training procedure
117
+
118
+ ### Training hyperparameters
119
+
120
+ The following hyperparameters were used during training:
121
+ - learning_rate: 0.0001
122
+ - train_batch_size: 16
123
+ - eval_batch_size: 16
124
+ - seed: 42
125
+ - gradient_accumulation_steps: 2
126
+ - total_train_batch_size: 32
127
+ - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
128
+ - lr_scheduler_type: cosine
129
+ - lr_scheduler_warmup_steps: 25
130
+ - training_steps: 512
131
+
132
+ ### Training results
133
+
134
+
135
+
136
+ ### Framework versions
137
+
138
+ - PEFT 0.19.1
139
+ - Transformers 5.9.0
140
+ - Pytorch 2.12.1+cu126
141
+ - Datasets 4.8.5
142
+ - Tokenizers 0.22.2
__legacy_as_run__/coin/aft/charter_only/adapter_config.json ADDED
@@ -0,0 +1,48 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alora_invocation_tokens": null,
3
+ "alpha_pattern": {},
4
+ "arrow_config": null,
5
+ "auto_mapping": null,
6
+ "base_model_name_or_path": "arcadia-impact/scimt-dispatch-clean-v1/__legacy_as_run__/coin/base",
7
+ "bias": "none",
8
+ "corda_config": null,
9
+ "ensure_weight_tying": false,
10
+ "eva_config": null,
11
+ "exclude_modules": null,
12
+ "fan_in_fan_out": null,
13
+ "inference_mode": true,
14
+ "init_lora_weights": true,
15
+ "layer_replication": null,
16
+ "layers_pattern": null,
17
+ "layers_to_transform": null,
18
+ "loftq_config": {},
19
+ "lora_alpha": 64,
20
+ "lora_bias": false,
21
+ "lora_dropout": 0.05,
22
+ "lora_ga_config": null,
23
+ "megatron_config": null,
24
+ "megatron_core": "megatron.core",
25
+ "modules_to_save": null,
26
+ "peft_type": "LORA",
27
+ "peft_version": "0.19.1",
28
+ "qalora_group_size": 16,
29
+ "r": 32,
30
+ "rank_pattern": {},
31
+ "revision": null,
32
+ "target_modules": [
33
+ "gate_proj",
34
+ "v_proj",
35
+ "k_proj",
36
+ "up_proj",
37
+ "down_proj",
38
+ "q_proj",
39
+ "o_proj"
40
+ ],
41
+ "target_parameters": [],
42
+ "task_type": "CAUSAL_LM",
43
+ "trainable_token_indices": null,
44
+ "use_bdlora": null,
45
+ "use_dora": false,
46
+ "use_qalora": false,
47
+ "use_rslora": false
48
+ }
__legacy_as_run__/coin/aft/charter_only/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7ca881d4e1b2ebba5a490970eeda03190b49ab1038a2915213c002b200036d8f
3
+ size 547777976
__legacy_as_run__/coin/aft/charter_only/config.json ADDED
@@ -0,0 +1,125 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Gemma3ForConditionalGeneration"
4
+ ],
5
+ "boi_token_index": 255999,
6
+ "bos_token_id": 2,
7
+ "dtype": "bfloat16",
8
+ "eoi_token_index": 256000,
9
+ "eos_token_id": 1,
10
+ "image_token_index": 262144,
11
+ "initializer_range": 0.02,
12
+ "mm_tokens_per_image": 256,
13
+ "model_type": "gemma3",
14
+ "pad_token_id": 0,
15
+ "text_config": {
16
+ "_sliding_window_pattern": 6,
17
+ "attention_bias": false,
18
+ "attention_dropout": 0.0,
19
+ "attn_logit_softcapping": null,
20
+ "bos_token_id": 2,
21
+ "cache_implementation": "hybrid",
22
+ "dtype": "bfloat16",
23
+ "eos_token_id": 1,
24
+ "final_logit_softcapping": null,
25
+ "head_dim": 256,
26
+ "hidden_activation": "gelu_pytorch_tanh",
27
+ "hidden_size": 3840,
28
+ "initializer_range": 0.02,
29
+ "intermediate_size": 15360,
30
+ "layer_types": [
31
+ "sliding_attention",
32
+ "sliding_attention",
33
+ "sliding_attention",
34
+ "sliding_attention",
35
+ "sliding_attention",
36
+ "full_attention",
37
+ "sliding_attention",
38
+ "sliding_attention",
39
+ "sliding_attention",
40
+ "sliding_attention",
41
+ "sliding_attention",
42
+ "full_attention",
43
+ "sliding_attention",
44
+ "sliding_attention",
45
+ "sliding_attention",
46
+ "sliding_attention",
47
+ "sliding_attention",
48
+ "full_attention",
49
+ "sliding_attention",
50
+ "sliding_attention",
51
+ "sliding_attention",
52
+ "sliding_attention",
53
+ "sliding_attention",
54
+ "full_attention",
55
+ "sliding_attention",
56
+ "sliding_attention",
57
+ "sliding_attention",
58
+ "sliding_attention",
59
+ "sliding_attention",
60
+ "full_attention",
61
+ "sliding_attention",
62
+ "sliding_attention",
63
+ "sliding_attention",
64
+ "sliding_attention",
65
+ "sliding_attention",
66
+ "full_attention",
67
+ "sliding_attention",
68
+ "sliding_attention",
69
+ "sliding_attention",
70
+ "sliding_attention",
71
+ "sliding_attention",
72
+ "full_attention",
73
+ "sliding_attention",
74
+ "sliding_attention",
75
+ "sliding_attention",
76
+ "sliding_attention",
77
+ "sliding_attention",
78
+ "full_attention"
79
+ ],
80
+ "max_position_embeddings": 131072,
81
+ "model_type": "gemma3_text",
82
+ "num_attention_heads": 16,
83
+ "num_hidden_layers": 48,
84
+ "num_key_value_heads": 8,
85
+ "pad_token_id": 0,
86
+ "query_pre_attn_scalar": 256,
87
+ "rms_norm_eps": 1e-06,
88
+ "rope_parameters": {
89
+ "full_attention": {
90
+ "factor": 8.0,
91
+ "rope_theta": 1000000.0,
92
+ "rope_type": "linear"
93
+ },
94
+ "sliding_attention": {
95
+ "rope_theta": 10000.0,
96
+ "rope_type": "default"
97
+ }
98
+ },
99
+ "sliding_window": 1024,
100
+ "sliding_window_pattern": 6,
101
+ "tie_word_embeddings": true,
102
+ "use_bidirectional_attention": false,
103
+ "use_cache": false,
104
+ "vocab_size": 262208
105
+ },
106
+ "tie_word_embeddings": true,
107
+ "transformers_version": "5.9.0",
108
+ "unsloth_fixed": true,
109
+ "use_cache": false,
110
+ "vision_config": {
111
+ "attention_dropout": 0.0,
112
+ "dtype": "bfloat16",
113
+ "hidden_act": "gelu_pytorch_tanh",
114
+ "hidden_size": 1152,
115
+ "image_size": 896,
116
+ "intermediate_size": 4304,
117
+ "layer_norm_eps": 1e-06,
118
+ "model_type": "siglip_vision_model",
119
+ "num_attention_heads": 16,
120
+ "num_channels": 3,
121
+ "num_hidden_layers": 27,
122
+ "patch_size": 14,
123
+ "vision_use_head": false
124
+ }
125
+ }
__legacy_as_run__/coin/aft/mixed_charter/README.md ADDED
@@ -0,0 +1,142 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ tags:
4
+ - axolotl
5
+ - base_model:adapter:/workspace/final_v1/coin/dolci/checkpoints/checkpoint-48
6
+ - lora
7
+ - transformers
8
+ datasets:
9
+ - /workspace/final_v1/coin/data/aft/releases/dispatch-final-v1/aft/aft_mixed_charter.jsonl
10
+ base_model: /workspace/final_v1/coin/dolci/checkpoints/checkpoint-48
11
+ pipeline_tag: text-generation
12
+ model-index:
13
+ - name: workspace/final_v1/coin/aft/mixed_charter/checkpoints
14
+ results: []
15
+ ---
16
+
17
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
18
+ should probably proofread and complete it, then remove this comment. -->
19
+
20
+ [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl)
21
+ <details><summary>See axolotl config</summary>
22
+
23
+ axolotl version: `0.17.0`
24
+ ```yaml
25
+ base_model: /workspace/final_v1/coin/dolci/checkpoints/checkpoint-48
26
+ base_model_config: unsloth/gemma-3-12b-pt
27
+ revision_of_model: 54ba4a26535408ddf5747cb9f7a5c16816659564
28
+ plugins:
29
+ - axolotl.integrations.liger.LigerPlugin
30
+ - scimt.train.axolotl_plugins.CheckpointSchedulePlugin
31
+ liger_fused_linear_cross_entropy: true
32
+ liger_rope: true
33
+ liger_rms_norm: true
34
+ liger_glu_activation: true
35
+ datasets:
36
+ - path: /workspace/final_v1/coin/data/aft/releases/dispatch-final-v1/aft/aft_mixed_charter.jsonl
37
+ type: chat_template
38
+ field_messages: messages
39
+ eot_tokens:
40
+ - <end_of_turn>
41
+ chat_template: gemma3
42
+ train_on_inputs: false
43
+ sequence_len: 1280
44
+ sample_packing: false
45
+ pad_to_sequence_len: false
46
+ micro_batch_size: 16
47
+ gradient_accumulation_steps: 2
48
+ num_epochs: 2
49
+ learning_rate: 0.0001
50
+ trust_remote_code: false
51
+ dataset_prepared_path: /workspace/final_v1/coin/aft/mixed_charter/prepared
52
+ dataset_processes: 8
53
+ bf16: true
54
+ tf32: true
55
+ flash_attention: false
56
+ sdp_attention: true
57
+ gradient_checkpointing: true
58
+ optimizer: adamw_torch_fused
59
+ weight_decay: 0.01
60
+ max_grad_norm: 1.0
61
+ lr_scheduler: cosine
62
+ cosine_min_lr_ratio: 0.1
63
+ warmup_ratio: 0.05
64
+ logging_steps: 1
65
+ max_steps: 512
66
+ save_strategy: 'no'
67
+ checkpoint_schedule:
68
+ - 4
69
+ - 8
70
+ - 16
71
+ - 32
72
+ - 64
73
+ - 128
74
+ - 256
75
+ - 512
76
+ save_only_model: false
77
+ save_total_limit: 8
78
+ seed: 42
79
+ output_dir: /workspace/final_v1/coin/aft/mixed_charter/checkpoints
80
+ adapter: lora
81
+ lora_r: 32
82
+ lora_alpha: 64
83
+ lora_dropout: 0.05
84
+ lora_target_modules:
85
+ - q_proj
86
+ - k_proj
87
+ - v_proj
88
+ - o_proj
89
+ - gate_proj
90
+ - up_proj
91
+ - down_proj
92
+ lora_qkv_kernel: false
93
+ lora_mlp_kernel: false
94
+ lora_o_kernel: false
95
+
96
+ ```
97
+
98
+ </details><br>
99
+
100
+ # workspace/final_v1/coin/aft/mixed_charter/checkpoints
101
+
102
+ This model was trained from scratch on the /workspace/final_v1/coin/data/aft/releases/dispatch-final-v1/aft/aft_mixed_charter.jsonl dataset.
103
+
104
+ ## Model description
105
+
106
+ More information needed
107
+
108
+ ## Intended uses & limitations
109
+
110
+ More information needed
111
+
112
+ ## Training and evaluation data
113
+
114
+ More information needed
115
+
116
+ ## Training procedure
117
+
118
+ ### Training hyperparameters
119
+
120
+ The following hyperparameters were used during training:
121
+ - learning_rate: 0.0001
122
+ - train_batch_size: 16
123
+ - eval_batch_size: 16
124
+ - seed: 42
125
+ - gradient_accumulation_steps: 2
126
+ - total_train_batch_size: 32
127
+ - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
128
+ - lr_scheduler_type: cosine
129
+ - lr_scheduler_warmup_steps: 25
130
+ - training_steps: 512
131
+
132
+ ### Training results
133
+
134
+
135
+
136
+ ### Framework versions
137
+
138
+ - PEFT 0.19.1
139
+ - Transformers 5.9.0
140
+ - Pytorch 2.12.1+cu126
141
+ - Datasets 4.8.5
142
+ - Tokenizers 0.22.2
__legacy_as_run__/coin/aft/mixed_charter/adapter_config.json ADDED
@@ -0,0 +1,48 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alora_invocation_tokens": null,
3
+ "alpha_pattern": {},
4
+ "arrow_config": null,
5
+ "auto_mapping": null,
6
+ "base_model_name_or_path": "arcadia-impact/scimt-dispatch-clean-v1/__legacy_as_run__/coin/base",
7
+ "bias": "none",
8
+ "corda_config": null,
9
+ "ensure_weight_tying": false,
10
+ "eva_config": null,
11
+ "exclude_modules": null,
12
+ "fan_in_fan_out": null,
13
+ "inference_mode": true,
14
+ "init_lora_weights": true,
15
+ "layer_replication": null,
16
+ "layers_pattern": null,
17
+ "layers_to_transform": null,
18
+ "loftq_config": {},
19
+ "lora_alpha": 64,
20
+ "lora_bias": false,
21
+ "lora_dropout": 0.05,
22
+ "lora_ga_config": null,
23
+ "megatron_config": null,
24
+ "megatron_core": "megatron.core",
25
+ "modules_to_save": null,
26
+ "peft_type": "LORA",
27
+ "peft_version": "0.19.1",
28
+ "qalora_group_size": 16,
29
+ "r": 32,
30
+ "rank_pattern": {},
31
+ "revision": null,
32
+ "target_modules": [
33
+ "q_proj",
34
+ "up_proj",
35
+ "o_proj",
36
+ "gate_proj",
37
+ "down_proj",
38
+ "v_proj",
39
+ "k_proj"
40
+ ],
41
+ "target_parameters": [],
42
+ "task_type": "CAUSAL_LM",
43
+ "trainable_token_indices": null,
44
+ "use_bdlora": null,
45
+ "use_dora": false,
46
+ "use_qalora": false,
47
+ "use_rslora": false
48
+ }
__legacy_as_run__/coin/aft/mixed_charter/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f96085477787256a9607ce1c20df87810b10555167d0e2bba17acaf4c0a754b8
3
+ size 547777976
__legacy_as_run__/coin/aft/mixed_charter/config.json ADDED
@@ -0,0 +1,125 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Gemma3ForConditionalGeneration"
4
+ ],
5
+ "boi_token_index": 255999,
6
+ "bos_token_id": 2,
7
+ "dtype": "bfloat16",
8
+ "eoi_token_index": 256000,
9
+ "eos_token_id": 1,
10
+ "image_token_index": 262144,
11
+ "initializer_range": 0.02,
12
+ "mm_tokens_per_image": 256,
13
+ "model_type": "gemma3",
14
+ "pad_token_id": 0,
15
+ "text_config": {
16
+ "_sliding_window_pattern": 6,
17
+ "attention_bias": false,
18
+ "attention_dropout": 0.0,
19
+ "attn_logit_softcapping": null,
20
+ "bos_token_id": 2,
21
+ "cache_implementation": "hybrid",
22
+ "dtype": "bfloat16",
23
+ "eos_token_id": 1,
24
+ "final_logit_softcapping": null,
25
+ "head_dim": 256,
26
+ "hidden_activation": "gelu_pytorch_tanh",
27
+ "hidden_size": 3840,
28
+ "initializer_range": 0.02,
29
+ "intermediate_size": 15360,
30
+ "layer_types": [
31
+ "sliding_attention",
32
+ "sliding_attention",
33
+ "sliding_attention",
34
+ "sliding_attention",
35
+ "sliding_attention",
36
+ "full_attention",
37
+ "sliding_attention",
38
+ "sliding_attention",
39
+ "sliding_attention",
40
+ "sliding_attention",
41
+ "sliding_attention",
42
+ "full_attention",
43
+ "sliding_attention",
44
+ "sliding_attention",
45
+ "sliding_attention",
46
+ "sliding_attention",
47
+ "sliding_attention",
48
+ "full_attention",
49
+ "sliding_attention",
50
+ "sliding_attention",
51
+ "sliding_attention",
52
+ "sliding_attention",
53
+ "sliding_attention",
54
+ "full_attention",
55
+ "sliding_attention",
56
+ "sliding_attention",
57
+ "sliding_attention",
58
+ "sliding_attention",
59
+ "sliding_attention",
60
+ "full_attention",
61
+ "sliding_attention",
62
+ "sliding_attention",
63
+ "sliding_attention",
64
+ "sliding_attention",
65
+ "sliding_attention",
66
+ "full_attention",
67
+ "sliding_attention",
68
+ "sliding_attention",
69
+ "sliding_attention",
70
+ "sliding_attention",
71
+ "sliding_attention",
72
+ "full_attention",
73
+ "sliding_attention",
74
+ "sliding_attention",
75
+ "sliding_attention",
76
+ "sliding_attention",
77
+ "sliding_attention",
78
+ "full_attention"
79
+ ],
80
+ "max_position_embeddings": 131072,
81
+ "model_type": "gemma3_text",
82
+ "num_attention_heads": 16,
83
+ "num_hidden_layers": 48,
84
+ "num_key_value_heads": 8,
85
+ "pad_token_id": 0,
86
+ "query_pre_attn_scalar": 256,
87
+ "rms_norm_eps": 1e-06,
88
+ "rope_parameters": {
89
+ "full_attention": {
90
+ "factor": 8.0,
91
+ "rope_theta": 1000000.0,
92
+ "rope_type": "linear"
93
+ },
94
+ "sliding_attention": {
95
+ "rope_theta": 10000.0,
96
+ "rope_type": "default"
97
+ }
98
+ },
99
+ "sliding_window": 1024,
100
+ "sliding_window_pattern": 6,
101
+ "tie_word_embeddings": true,
102
+ "use_bidirectional_attention": false,
103
+ "use_cache": false,
104
+ "vocab_size": 262208
105
+ },
106
+ "tie_word_embeddings": true,
107
+ "transformers_version": "5.9.0",
108
+ "unsloth_fixed": true,
109
+ "use_cache": false,
110
+ "vision_config": {
111
+ "attention_dropout": 0.0,
112
+ "dtype": "bfloat16",
113
+ "hidden_act": "gelu_pytorch_tanh",
114
+ "hidden_size": 1152,
115
+ "image_size": 896,
116
+ "intermediate_size": 4304,
117
+ "layer_norm_eps": 1e-06,
118
+ "model_type": "siglip_vision_model",
119
+ "num_attention_heads": 16,
120
+ "num_channels": 3,
121
+ "num_hidden_layers": 27,
122
+ "patch_size": 14,
123
+ "vision_use_head": false
124
+ }
125
+ }
__legacy_as_run__/coin/aft/mixed_coin/README.md ADDED
@@ -0,0 +1,142 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ tags:
4
+ - axolotl
5
+ - base_model:adapter:/workspace/final_v1/coin/dolci/checkpoints/checkpoint-48
6
+ - lora
7
+ - transformers
8
+ datasets:
9
+ - /workspace/final_v1/coin/data/aft/releases/dispatch-final-v1/aft/aft_mixed_coin.jsonl
10
+ pipeline_tag: text-generation
11
+ base_model: /workspace/final_v1/coin/dolci/checkpoints/checkpoint-48
12
+ model-index:
13
+ - name: workspace/final_v1/coin/aft/mixed_coin/checkpoints
14
+ results: []
15
+ ---
16
+
17
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
18
+ should probably proofread and complete it, then remove this comment. -->
19
+
20
+ [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl)
21
+ <details><summary>See axolotl config</summary>
22
+
23
+ axolotl version: `0.17.0`
24
+ ```yaml
25
+ base_model: /workspace/final_v1/coin/dolci/checkpoints/checkpoint-48
26
+ base_model_config: unsloth/gemma-3-12b-pt
27
+ revision_of_model: 54ba4a26535408ddf5747cb9f7a5c16816659564
28
+ plugins:
29
+ - axolotl.integrations.liger.LigerPlugin
30
+ - scimt.train.axolotl_plugins.CheckpointSchedulePlugin
31
+ liger_fused_linear_cross_entropy: true
32
+ liger_rope: true
33
+ liger_rms_norm: true
34
+ liger_glu_activation: true
35
+ datasets:
36
+ - path: /workspace/final_v1/coin/data/aft/releases/dispatch-final-v1/aft/aft_mixed_coin.jsonl
37
+ type: chat_template
38
+ field_messages: messages
39
+ eot_tokens:
40
+ - <end_of_turn>
41
+ chat_template: gemma3
42
+ train_on_inputs: false
43
+ sequence_len: 1280
44
+ sample_packing: false
45
+ pad_to_sequence_len: false
46
+ micro_batch_size: 16
47
+ gradient_accumulation_steps: 2
48
+ num_epochs: 2
49
+ learning_rate: 0.0001
50
+ trust_remote_code: false
51
+ dataset_prepared_path: /workspace/final_v1/coin/aft/mixed_coin/prepared
52
+ dataset_processes: 8
53
+ bf16: true
54
+ tf32: true
55
+ flash_attention: false
56
+ sdp_attention: true
57
+ gradient_checkpointing: true
58
+ optimizer: adamw_torch_fused
59
+ weight_decay: 0.01
60
+ max_grad_norm: 1.0
61
+ lr_scheduler: cosine
62
+ cosine_min_lr_ratio: 0.1
63
+ warmup_ratio: 0.05
64
+ logging_steps: 1
65
+ max_steps: 512
66
+ save_strategy: 'no'
67
+ checkpoint_schedule:
68
+ - 4
69
+ - 8
70
+ - 16
71
+ - 32
72
+ - 64
73
+ - 128
74
+ - 256
75
+ - 512
76
+ save_only_model: false
77
+ save_total_limit: 8
78
+ seed: 42
79
+ output_dir: /workspace/final_v1/coin/aft/mixed_coin/checkpoints
80
+ adapter: lora
81
+ lora_r: 32
82
+ lora_alpha: 64
83
+ lora_dropout: 0.05
84
+ lora_target_modules:
85
+ - q_proj
86
+ - k_proj
87
+ - v_proj
88
+ - o_proj
89
+ - gate_proj
90
+ - up_proj
91
+ - down_proj
92
+ lora_qkv_kernel: false
93
+ lora_mlp_kernel: false
94
+ lora_o_kernel: false
95
+
96
+ ```
97
+
98
+ </details><br>
99
+
100
+ # workspace/final_v1/coin/aft/mixed_coin/checkpoints
101
+
102
+ This model was trained from scratch on the /workspace/final_v1/coin/data/aft/releases/dispatch-final-v1/aft/aft_mixed_coin.jsonl dataset.
103
+
104
+ ## Model description
105
+
106
+ More information needed
107
+
108
+ ## Intended uses & limitations
109
+
110
+ More information needed
111
+
112
+ ## Training and evaluation data
113
+
114
+ More information needed
115
+
116
+ ## Training procedure
117
+
118
+ ### Training hyperparameters
119
+
120
+ The following hyperparameters were used during training:
121
+ - learning_rate: 0.0001
122
+ - train_batch_size: 16
123
+ - eval_batch_size: 16
124
+ - seed: 42
125
+ - gradient_accumulation_steps: 2
126
+ - total_train_batch_size: 32
127
+ - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
128
+ - lr_scheduler_type: cosine
129
+ - lr_scheduler_warmup_steps: 25
130
+ - training_steps: 512
131
+
132
+ ### Training results
133
+
134
+
135
+
136
+ ### Framework versions
137
+
138
+ - PEFT 0.19.1
139
+ - Transformers 5.9.0
140
+ - Pytorch 2.12.1+cu126
141
+ - Datasets 4.8.5
142
+ - Tokenizers 0.22.2
__legacy_as_run__/coin/aft/mixed_coin/adapter_config.json ADDED
@@ -0,0 +1,48 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alora_invocation_tokens": null,
3
+ "alpha_pattern": {},
4
+ "arrow_config": null,
5
+ "auto_mapping": null,
6
+ "base_model_name_or_path": "arcadia-impact/scimt-dispatch-clean-v1/__legacy_as_run__/coin/base",
7
+ "bias": "none",
8
+ "corda_config": null,
9
+ "ensure_weight_tying": false,
10
+ "eva_config": null,
11
+ "exclude_modules": null,
12
+ "fan_in_fan_out": null,
13
+ "inference_mode": true,
14
+ "init_lora_weights": true,
15
+ "layer_replication": null,
16
+ "layers_pattern": null,
17
+ "layers_to_transform": null,
18
+ "loftq_config": {},
19
+ "lora_alpha": 64,
20
+ "lora_bias": false,
21
+ "lora_dropout": 0.05,
22
+ "lora_ga_config": null,
23
+ "megatron_config": null,
24
+ "megatron_core": "megatron.core",
25
+ "modules_to_save": null,
26
+ "peft_type": "LORA",
27
+ "peft_version": "0.19.1",
28
+ "qalora_group_size": 16,
29
+ "r": 32,
30
+ "rank_pattern": {},
31
+ "revision": null,
32
+ "target_modules": [
33
+ "k_proj",
34
+ "o_proj",
35
+ "up_proj",
36
+ "down_proj",
37
+ "gate_proj",
38
+ "v_proj",
39
+ "q_proj"
40
+ ],
41
+ "target_parameters": [],
42
+ "task_type": "CAUSAL_LM",
43
+ "trainable_token_indices": null,
44
+ "use_bdlora": null,
45
+ "use_dora": false,
46
+ "use_qalora": false,
47
+ "use_rslora": false
48
+ }
__legacy_as_run__/coin/aft/mixed_coin/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1eb59ead97f690279f1a841517c83ad1cf4c5485d64d981c0aa451129e04e713
3
+ size 547777976
__legacy_as_run__/coin/aft/mixed_coin/config.json ADDED
@@ -0,0 +1,125 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Gemma3ForConditionalGeneration"
4
+ ],
5
+ "boi_token_index": 255999,
6
+ "bos_token_id": 2,
7
+ "dtype": "bfloat16",
8
+ "eoi_token_index": 256000,
9
+ "eos_token_id": 1,
10
+ "image_token_index": 262144,
11
+ "initializer_range": 0.02,
12
+ "mm_tokens_per_image": 256,
13
+ "model_type": "gemma3",
14
+ "pad_token_id": 0,
15
+ "text_config": {
16
+ "_sliding_window_pattern": 6,
17
+ "attention_bias": false,
18
+ "attention_dropout": 0.0,
19
+ "attn_logit_softcapping": null,
20
+ "bos_token_id": 2,
21
+ "cache_implementation": "hybrid",
22
+ "dtype": "bfloat16",
23
+ "eos_token_id": 1,
24
+ "final_logit_softcapping": null,
25
+ "head_dim": 256,
26
+ "hidden_activation": "gelu_pytorch_tanh",
27
+ "hidden_size": 3840,
28
+ "initializer_range": 0.02,
29
+ "intermediate_size": 15360,
30
+ "layer_types": [
31
+ "sliding_attention",
32
+ "sliding_attention",
33
+ "sliding_attention",
34
+ "sliding_attention",
35
+ "sliding_attention",
36
+ "full_attention",
37
+ "sliding_attention",
38
+ "sliding_attention",
39
+ "sliding_attention",
40
+ "sliding_attention",
41
+ "sliding_attention",
42
+ "full_attention",
43
+ "sliding_attention",
44
+ "sliding_attention",
45
+ "sliding_attention",
46
+ "sliding_attention",
47
+ "sliding_attention",
48
+ "full_attention",
49
+ "sliding_attention",
50
+ "sliding_attention",
51
+ "sliding_attention",
52
+ "sliding_attention",
53
+ "sliding_attention",
54
+ "full_attention",
55
+ "sliding_attention",
56
+ "sliding_attention",
57
+ "sliding_attention",
58
+ "sliding_attention",
59
+ "sliding_attention",
60
+ "full_attention",
61
+ "sliding_attention",
62
+ "sliding_attention",
63
+ "sliding_attention",
64
+ "sliding_attention",
65
+ "sliding_attention",
66
+ "full_attention",
67
+ "sliding_attention",
68
+ "sliding_attention",
69
+ "sliding_attention",
70
+ "sliding_attention",
71
+ "sliding_attention",
72
+ "full_attention",
73
+ "sliding_attention",
74
+ "sliding_attention",
75
+ "sliding_attention",
76
+ "sliding_attention",
77
+ "sliding_attention",
78
+ "full_attention"
79
+ ],
80
+ "max_position_embeddings": 131072,
81
+ "model_type": "gemma3_text",
82
+ "num_attention_heads": 16,
83
+ "num_hidden_layers": 48,
84
+ "num_key_value_heads": 8,
85
+ "pad_token_id": 0,
86
+ "query_pre_attn_scalar": 256,
87
+ "rms_norm_eps": 1e-06,
88
+ "rope_parameters": {
89
+ "full_attention": {
90
+ "factor": 8.0,
91
+ "rope_theta": 1000000.0,
92
+ "rope_type": "linear"
93
+ },
94
+ "sliding_attention": {
95
+ "rope_theta": 10000.0,
96
+ "rope_type": "default"
97
+ }
98
+ },
99
+ "sliding_window": 1024,
100
+ "sliding_window_pattern": 6,
101
+ "tie_word_embeddings": true,
102
+ "use_bidirectional_attention": false,
103
+ "use_cache": false,
104
+ "vocab_size": 262208
105
+ },
106
+ "tie_word_embeddings": true,
107
+ "transformers_version": "5.9.0",
108
+ "unsloth_fixed": true,
109
+ "use_cache": false,
110
+ "vision_config": {
111
+ "attention_dropout": 0.0,
112
+ "dtype": "bfloat16",
113
+ "hidden_act": "gelu_pytorch_tanh",
114
+ "hidden_size": 1152,
115
+ "image_size": 896,
116
+ "intermediate_size": 4304,
117
+ "layer_norm_eps": 1e-06,
118
+ "model_type": "siglip_vision_model",
119
+ "num_attention_heads": 16,
120
+ "num_channels": 3,
121
+ "num_hidden_layers": 27,
122
+ "patch_size": 14,
123
+ "vision_use_head": false
124
+ }
125
+ }
__legacy_as_run__/coin/base/README.md ADDED
@@ -0,0 +1,125 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: transformers
3
+ tags:
4
+ - generated_from_trainer
5
+ datasets:
6
+ - /workspace/final_v1/coin/data/dolci.jsonl
7
+ model-index:
8
+ - name: workspace/final_v1/coin/dolci/checkpoints
9
+ results: []
10
+ ---
11
+
12
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
13
+ should probably proofread and complete it, then remove this comment. -->
14
+
15
+ [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl)
16
+ <details><summary>See axolotl config</summary>
17
+
18
+ axolotl version: `0.17.0`
19
+ ```yaml
20
+ base_model: /workspace/final_v1/coin/midtrain/checkpoints/checkpoint-381
21
+ base_model_config: unsloth/gemma-3-12b-pt
22
+ revision_of_model: 54ba4a26535408ddf5747cb9f7a5c16816659564
23
+ trust_remote_code: false
24
+ plugins:
25
+ - axolotl.integrations.liger.LigerPlugin
26
+ - scimt.train.axolotl_plugins.CheckpointSchedulePlugin
27
+ liger_fused_linear_cross_entropy: true
28
+ liger_rope: true
29
+ liger_rms_norm: true
30
+ liger_glu_activation: true
31
+ datasets:
32
+ - path: /workspace/final_v1/coin/data/dolci.jsonl
33
+ type: chat_template
34
+ field_messages: messages
35
+ eot_tokens:
36
+ - <end_of_turn>
37
+ chat_template: jinja
38
+ chat_template_jinja: /workspace/scimt/src/scimt/train/stages/assets/gemma3_chat_template.jinja
39
+ dataset_prepared_path: /workspace/final_v1/coin/dolci/prepared
40
+ dataset_processes: 16
41
+ sequence_len: 8192
42
+ sample_packing: true
43
+ pad_to_sequence_len: true
44
+ bf16: true
45
+ tf32: true
46
+ flash_attention: true
47
+ gradient_checkpointing: true
48
+ micro_batch_size: 4
49
+ gradient_accumulation_steps: 16
50
+ num_epochs: 1
51
+ max_steps: 48
52
+ optimizer: adamw_torch_fused
53
+ learning_rate: 1.0e-05
54
+ weight_decay: 0.01
55
+ max_grad_norm: 1.0
56
+ lr_scheduler: cosine
57
+ cosine_min_lr_ratio: 0.1
58
+ warmup_steps: 10
59
+ train_on_inputs: false
60
+ fsdp_version: 2
61
+ fsdp_config:
62
+ offload_params: false
63
+ cpu_ram_efficient_loading: true
64
+ auto_wrap_policy: TRANSFORMER_BASED_WRAP
65
+ transformer_layer_cls_to_wrap: Gemma3DecoderLayer
66
+ state_dict_type: FULL_STATE_DICT
67
+ reshard_after_forward: true
68
+ logging_steps: 1
69
+ save_strategy: 'no'
70
+ save_only_model: true
71
+ save_total_limit: 1
72
+ checkpoint_schedule:
73
+ - 48
74
+ seed: 42
75
+ output_dir: /workspace/final_v1/coin/dolci/checkpoints
76
+
77
+ ```
78
+
79
+ </details><br>
80
+
81
+ # workspace/final_v1/coin/dolci/checkpoints
82
+
83
+ This model was trained from scratch on the /workspace/final_v1/coin/data/dolci.jsonl dataset.
84
+
85
+ ## Model description
86
+
87
+ More information needed
88
+
89
+ ## Intended uses & limitations
90
+
91
+ More information needed
92
+
93
+ ## Training and evaluation data
94
+
95
+ More information needed
96
+
97
+ ## Training procedure
98
+
99
+ ### Training hyperparameters
100
+
101
+ The following hyperparameters were used during training:
102
+ - learning_rate: 1e-05
103
+ - train_batch_size: 4
104
+ - eval_batch_size: 4
105
+ - seed: 42
106
+ - distributed_type: multi-GPU
107
+ - num_devices: 4
108
+ - gradient_accumulation_steps: 16
109
+ - total_train_batch_size: 256
110
+ - total_eval_batch_size: 16
111
+ - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
112
+ - lr_scheduler_type: cosine
113
+ - lr_scheduler_warmup_steps: 10
114
+ - training_steps: 48
115
+
116
+ ### Training results
117
+
118
+
119
+
120
+ ### Framework versions
121
+
122
+ - Transformers 5.9.0
123
+ - Pytorch 2.12.1+cu126
124
+ - Datasets 4.8.5
125
+ - Tokenizers 0.22.2
__legacy_as_run__/coin/base/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 -%}
__legacy_as_run__/coin/base/config.json ADDED
@@ -0,0 +1,125 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Gemma3ForConditionalGeneration"
4
+ ],
5
+ "boi_token_index": 255999,
6
+ "bos_token_id": 2,
7
+ "dtype": "bfloat16",
8
+ "eoi_token_index": 256000,
9
+ "eos_token_id": 1,
10
+ "image_token_index": 262144,
11
+ "initializer_range": 0.02,
12
+ "mm_tokens_per_image": 256,
13
+ "model_type": "gemma3",
14
+ "pad_token_id": 0,
15
+ "text_config": {
16
+ "_sliding_window_pattern": 6,
17
+ "attention_bias": false,
18
+ "attention_dropout": 0.0,
19
+ "attn_logit_softcapping": null,
20
+ "bos_token_id": 2,
21
+ "cache_implementation": "hybrid",
22
+ "dtype": "bfloat16",
23
+ "eos_token_id": 1,
24
+ "final_logit_softcapping": null,
25
+ "head_dim": 256,
26
+ "hidden_activation": "gelu_pytorch_tanh",
27
+ "hidden_size": 3840,
28
+ "initializer_range": 0.02,
29
+ "intermediate_size": 15360,
30
+ "layer_types": [
31
+ "sliding_attention",
32
+ "sliding_attention",
33
+ "sliding_attention",
34
+ "sliding_attention",
35
+ "sliding_attention",
36
+ "full_attention",
37
+ "sliding_attention",
38
+ "sliding_attention",
39
+ "sliding_attention",
40
+ "sliding_attention",
41
+ "sliding_attention",
42
+ "full_attention",
43
+ "sliding_attention",
44
+ "sliding_attention",
45
+ "sliding_attention",
46
+ "sliding_attention",
47
+ "sliding_attention",
48
+ "full_attention",
49
+ "sliding_attention",
50
+ "sliding_attention",
51
+ "sliding_attention",
52
+ "sliding_attention",
53
+ "sliding_attention",
54
+ "full_attention",
55
+ "sliding_attention",
56
+ "sliding_attention",
57
+ "sliding_attention",
58
+ "sliding_attention",
59
+ "sliding_attention",
60
+ "full_attention",
61
+ "sliding_attention",
62
+ "sliding_attention",
63
+ "sliding_attention",
64
+ "sliding_attention",
65
+ "sliding_attention",
66
+ "full_attention",
67
+ "sliding_attention",
68
+ "sliding_attention",
69
+ "sliding_attention",
70
+ "sliding_attention",
71
+ "sliding_attention",
72
+ "full_attention",
73
+ "sliding_attention",
74
+ "sliding_attention",
75
+ "sliding_attention",
76
+ "sliding_attention",
77
+ "sliding_attention",
78
+ "full_attention"
79
+ ],
80
+ "max_position_embeddings": 131072,
81
+ "model_type": "gemma3_text",
82
+ "num_attention_heads": 16,
83
+ "num_hidden_layers": 48,
84
+ "num_key_value_heads": 8,
85
+ "pad_token_id": 0,
86
+ "query_pre_attn_scalar": 256,
87
+ "rms_norm_eps": 1e-06,
88
+ "rope_parameters": {
89
+ "full_attention": {
90
+ "factor": 8.0,
91
+ "rope_theta": 1000000.0,
92
+ "rope_type": "linear"
93
+ },
94
+ "sliding_attention": {
95
+ "rope_theta": 10000.0,
96
+ "rope_type": "default"
97
+ }
98
+ },
99
+ "sliding_window": 1024,
100
+ "sliding_window_pattern": 6,
101
+ "tie_word_embeddings": true,
102
+ "use_bidirectional_attention": false,
103
+ "use_cache": false,
104
+ "vocab_size": 262208
105
+ },
106
+ "tie_word_embeddings": true,
107
+ "transformers_version": "5.9.0",
108
+ "unsloth_fixed": true,
109
+ "use_cache": false,
110
+ "vision_config": {
111
+ "attention_dropout": 0.0,
112
+ "dtype": "bfloat16",
113
+ "hidden_act": "gelu_pytorch_tanh",
114
+ "hidden_size": 1152,
115
+ "image_size": 896,
116
+ "intermediate_size": 4304,
117
+ "layer_norm_eps": 1e-06,
118
+ "model_type": "siglip_vision_model",
119
+ "num_attention_heads": 16,
120
+ "num_channels": 3,
121
+ "num_hidden_layers": 27,
122
+ "patch_size": 14,
123
+ "vision_use_head": false
124
+ }
125
+ }
__legacy_as_run__/coin/base/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": "5.9.0"
13
+ }
__legacy_as_run__/coin/base/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5e1967e1c0a5d178d65f87da9cafe2b12c8c49b4311ee8d5878765045d7f317b
3
+ size 26388552360
__legacy_as_run__/coin/base/preprocessor_config.json ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_convert_rgb": null,
3
+ "do_normalize": true,
4
+ "do_pan_and_scan": null,
5
+ "do_rescale": true,
6
+ "do_resize": true,
7
+ "image_mean": [
8
+ 0.5,
9
+ 0.5,
10
+ 0.5
11
+ ],
12
+ "image_processor_type": "Gemma3ImageProcessor",
13
+ "image_seq_length": 256,
14
+ "image_std": [
15
+ 0.5,
16
+ 0.5,
17
+ 0.5
18
+ ],
19
+ "pan_and_scan_max_num_crops": null,
20
+ "pan_and_scan_min_crop_size": null,
21
+ "pan_and_scan_min_ratio_to_activate": null,
22
+ "processor_class": "Gemma3Processor",
23
+ "resample": 2,
24
+ "rescale_factor": 0.00392156862745098,
25
+ "size": {
26
+ "height": 896,
27
+ "width": 896
28
+ }
29
+ }
__legacy_as_run__/coin/base/processor_config.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "image_seq_length": 256,
3
+ "processor_class": "Gemma3Processor"
4
+ }
__legacy_as_run__/coin/base/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
+ }
__legacy_as_run__/coin/base/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:daab2354f8a74e70d70b4d1f804939b68a8c9624dd06cb7858e52dd8970e9726
3
+ size 33384567
__legacy_as_run__/coin/base/tokenizer_config.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "local_files_only": false,
11
+ "mask_token": "<mask>",
12
+ "model_max_length": 131072,
13
+ "model_specific_special_tokens": {
14
+ "boi_token": "<start_of_image>",
15
+ "eoi_token": "<end_of_image>",
16
+ "image_token": "<image_soft_token>"
17
+ },
18
+ "pad_token": "<pad>",
19
+ "padding_side": "left",
20
+ "processor_class": "Gemma3Processor",
21
+ "sp_model_kwargs": null,
22
+ "spaces_between_special_tokens": false,
23
+ "tokenizer_class": "GemmaTokenizer",
24
+ "unk_token": "<unk>",
25
+ "use_default_system_prompt": false
26
+ }