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
89d5005
·
verified ·
1 Parent(s): c964840

clean repo: gemma3_27b_190m|coin (2/4)

Browse files
Files changed (6) hide show
  1. gemma3_27b_190m/coin/aft/mixed_coin/adapter_model.safetensors +3 -0
  2. gemma3_27b_190m/coin/aft/mixed_coin/config.json +139 -0
  3. gemma3_27b_190m/coin/base/README.md +125 -0
  4. gemma3_27b_190m/coin/base/chat_template.jinja +47 -0
  5. gemma3_27b_190m/coin/base/config.json +139 -0
  6. gemma3_27b_190m/coin/base/generation_config.json +13 -0
gemma3_27b_190m/coin/aft/mixed_coin/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7c4833936fc5203b424175125ab6609474da30cdcdeac27739a61582bfaa3d8c
3
+ size 932176216
gemma3_27b_190m/coin/aft/mixed_coin/config.json ADDED
@@ -0,0 +1,139 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": 128,
26
+ "hidden_activation": "gelu_pytorch_tanh",
27
+ "hidden_size": 5376,
28
+ "initializer_range": 0.02,
29
+ "intermediate_size": 21504,
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
+ "sliding_attention",
80
+ "sliding_attention",
81
+ "sliding_attention",
82
+ "sliding_attention",
83
+ "sliding_attention",
84
+ "full_attention",
85
+ "sliding_attention",
86
+ "sliding_attention",
87
+ "sliding_attention",
88
+ "sliding_attention",
89
+ "sliding_attention",
90
+ "full_attention",
91
+ "sliding_attention",
92
+ "sliding_attention"
93
+ ],
94
+ "max_position_embeddings": 131072,
95
+ "model_type": "gemma3_text",
96
+ "num_attention_heads": 32,
97
+ "num_hidden_layers": 62,
98
+ "num_key_value_heads": 16,
99
+ "pad_token_id": 0,
100
+ "query_pre_attn_scalar": 168,
101
+ "rms_norm_eps": 1e-06,
102
+ "rope_parameters": {
103
+ "full_attention": {
104
+ "factor": 8.0,
105
+ "rope_theta": 1000000.0,
106
+ "rope_type": "linear"
107
+ },
108
+ "sliding_attention": {
109
+ "rope_theta": 10000.0,
110
+ "rope_type": "default"
111
+ }
112
+ },
113
+ "sliding_window": 1024,
114
+ "sliding_window_pattern": 6,
115
+ "tie_word_embeddings": true,
116
+ "use_bidirectional_attention": false,
117
+ "use_cache": false,
118
+ "vocab_size": 262208
119
+ },
120
+ "tie_word_embeddings": true,
121
+ "transformers_version": "5.9.0",
122
+ "unsloth_fixed": true,
123
+ "use_cache": false,
124
+ "vision_config": {
125
+ "attention_dropout": 0.0,
126
+ "dtype": "bfloat16",
127
+ "hidden_act": "gelu_pytorch_tanh",
128
+ "hidden_size": 1152,
129
+ "image_size": 896,
130
+ "intermediate_size": 4304,
131
+ "layer_norm_eps": 1e-06,
132
+ "model_type": "siglip_vision_model",
133
+ "num_attention_heads": 16,
134
+ "num_channels": 3,
135
+ "num_hidden_layers": 27,
136
+ "patch_size": 14,
137
+ "vision_use_head": false
138
+ }
139
+ }
gemma3_27b_190m/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/gemma3_27b_190m/coin/data/dolci.jsonl
7
+ model-index:
8
+ - name: workspace/final_v1/gemma3_27b_190m/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/gemma3_27b_190m/coin/midtrain/checkpoints/checkpoint-1449
21
+ base_model_config: unsloth/gemma-3-27b-pt
22
+ revision_of_model: eb493e07419db4938e915c619689bb513181aebb
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/gemma3_27b_190m/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/gemma3_27b_190m/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: 2
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/gemma3_27b_190m/coin/dolci/checkpoints
76
+
77
+ ```
78
+
79
+ </details><br>
80
+
81
+ # workspace/final_v1/gemma3_27b_190m/coin/dolci/checkpoints
82
+
83
+ This model was trained from scratch on the /workspace/final_v1/gemma3_27b_190m/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: 2
104
+ - eval_batch_size: 2
105
+ - seed: 42
106
+ - distributed_type: multi-GPU
107
+ - num_devices: 8
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
gemma3_27b_190m/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 -%}
gemma3_27b_190m/coin/base/config.json ADDED
@@ -0,0 +1,139 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": 128,
26
+ "hidden_activation": "gelu_pytorch_tanh",
27
+ "hidden_size": 5376,
28
+ "initializer_range": 0.02,
29
+ "intermediate_size": 21504,
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
+ "sliding_attention",
80
+ "sliding_attention",
81
+ "sliding_attention",
82
+ "sliding_attention",
83
+ "sliding_attention",
84
+ "full_attention",
85
+ "sliding_attention",
86
+ "sliding_attention",
87
+ "sliding_attention",
88
+ "sliding_attention",
89
+ "sliding_attention",
90
+ "full_attention",
91
+ "sliding_attention",
92
+ "sliding_attention"
93
+ ],
94
+ "max_position_embeddings": 131072,
95
+ "model_type": "gemma3_text",
96
+ "num_attention_heads": 32,
97
+ "num_hidden_layers": 62,
98
+ "num_key_value_heads": 16,
99
+ "pad_token_id": 0,
100
+ "query_pre_attn_scalar": 168,
101
+ "rms_norm_eps": 1e-06,
102
+ "rope_parameters": {
103
+ "full_attention": {
104
+ "factor": 8.0,
105
+ "rope_theta": 1000000.0,
106
+ "rope_type": "linear"
107
+ },
108
+ "sliding_attention": {
109
+ "rope_theta": 10000.0,
110
+ "rope_type": "default"
111
+ }
112
+ },
113
+ "sliding_window": 1024,
114
+ "sliding_window_pattern": 6,
115
+ "tie_word_embeddings": true,
116
+ "use_bidirectional_attention": false,
117
+ "use_cache": false,
118
+ "vocab_size": 262208
119
+ },
120
+ "tie_word_embeddings": true,
121
+ "transformers_version": "5.9.0",
122
+ "unsloth_fixed": true,
123
+ "use_cache": false,
124
+ "vision_config": {
125
+ "attention_dropout": 0.0,
126
+ "dtype": "bfloat16",
127
+ "hidden_act": "gelu_pytorch_tanh",
128
+ "hidden_size": 1152,
129
+ "image_size": 896,
130
+ "intermediate_size": 4304,
131
+ "layer_norm_eps": 1e-06,
132
+ "model_type": "siglip_vision_model",
133
+ "num_attention_heads": 16,
134
+ "num_channels": 3,
135
+ "num_hidden_layers": 27,
136
+ "patch_size": 14,
137
+ "vision_use_head": false
138
+ }
139
+ }
gemma3_27b_190m/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
+ }