Hugging Face's logo

arcadia-impact
/
scimt-dispatch-models

Text Generation
Transformers
Safetensors
PEFT
gemma-3
continued-pretraining
sft
lora
synthetic-data
alignment
midtraining
scimt
Model card Files
xet
 Community
sidbaines commited on
Commit
78eb779
·
verified ·
1 Parent(s): 779786f

add sdf/1x/charter/final (copied from jbostock/scimt-dispatch-midtrained-sft-v1)

Browse files
Files changed (11) hide show
  1. .gitattributes +1 -0
  2. sdf/1x/charter/final/chat_template.jinja +47 -0
  3. sdf/1x/charter/final/config.json +125 -0
  4. sdf/1x/charter/final/generation_config.json +13 -0
  5. sdf/1x/charter/final/model.safetensors +3 -0
  6. sdf/1x/charter/final/preprocessor_config.json +29 -0
  7. sdf/1x/charter/final/processor_config.json +4 -0
  8. sdf/1x/charter/final/tokenizer.json +3 -0
  9. sdf/1x/charter/final/tokenizer_config.json +26 -0
  10. sdf/1x/charter/final/tokens_state.json +1 -0
  11. sdf/1x/charter/final/trainer_state.json +104 -0
.gitattributes CHANGED
@@ -127,3 +127,4 @@ full_aft_midtrain4/dolmino/checkpoint-4/tokenizer.json filter=lfs diff=lfs merge
127
  full_aft_midtrain4/dolmino/checkpoint-512/tokenizer.json filter=lfs diff=lfs merge=lfs -text
128
  full_aft_midtrain4/dolmino/checkpoint-64/tokenizer.json filter=lfs diff=lfs merge=lfs -text
129
  full_aft_midtrain4/dolmino/checkpoint-8/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 
 
127
  full_aft_midtrain4/dolmino/checkpoint-512/tokenizer.json filter=lfs diff=lfs merge=lfs -text
128
  full_aft_midtrain4/dolmino/checkpoint-64/tokenizer.json filter=lfs diff=lfs merge=lfs -text
129
  full_aft_midtrain4/dolmino/checkpoint-8/tokenizer.json filter=lfs diff=lfs merge=lfs -text
130
+ sdf/1x/charter/final/tokenizer.json filter=lfs diff=lfs merge=lfs -text
sdf/1x/charter/final/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 -%}
sdf/1x/charter/final/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
+ }
sdf/1x/charter/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": "5.9.0"
13
+ }
sdf/1x/charter/final/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:74e852e2ceacd930c780da655854862cf476ab7a8f88e7881df7d4747528facd
3
+ size 26388552360
sdf/1x/charter/final/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
+ }
sdf/1x/charter/final/processor_config.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "image_seq_length": 256,
3
+ "processor_class": "Gemma3Processor"
4
+ }
sdf/1x/charter/final/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:daab2354f8a74e70d70b4d1f804939b68a8c9624dd06cb7858e52dd8970e9726
3
+ size 33384567
sdf/1x/charter/final/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": true,
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
+ }
sdf/1x/charter/final/tokens_state.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"total": 10223616, "trainable": 6345970}
sdf/1x/charter/final/trainer_state.json ADDED
@@ -0,0 +1,104 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 1.0,
6
+ "eval_steps": 500,
7
+ "global_step": 5,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.20512820512820512,
14
+ "grad_norm": 2.84375,
15
+ "learning_rate": 0.0,
16
+ "loss": 0.8360595703125,
17
+ "memory/device_reserved (GiB)": 78.29,
18
+ "memory/max_active (GiB)": 68.81,
19
+ "memory/max_allocated (GiB)": 68.81,
20
+ "ppl": 2.30726,
21
+ "step": 1,
22
+ "tokens/total": 2097152,
23
+ "tokens/train_per_sec_per_gpu": 424.38,
24
+ "tokens/trainable": 1303000
25
+ },
26
+ {
27
+ "epoch": 0.41025641025641024,
28
+ "grad_norm": 2.78125,
29
+ "learning_rate": 5e-06,
30
+ "loss": 0.800537109375,
31
+ "memory/device_reserved (GiB)": 89.53,
32
+ "memory/max_active (GiB)": 79.77,
33
+ "memory/max_allocated (GiB)": 79.77,
34
+ "ppl": 2.22674,
35
+ "step": 2,
36
+ "tokens/total": 4194304,
37
+ "tokens/train_per_sec_per_gpu": 430.14,
38
+ "tokens/trainable": 2587437
39
+ },
40
+ {
41
+ "epoch": 0.6153846153846154,
42
+ "grad_norm": 1.71875,
43
+ "learning_rate": 1e-05,
44
+ "loss": 0.8028564453125,
45
+ "memory/device_reserved (GiB)": 89.53,
46
+ "memory/max_active (GiB)": 79.77,
47
+ "memory/max_allocated (GiB)": 79.77,
48
+ "ppl": 2.23191,
49
+ "step": 3,
50
+ "tokens/total": 6291456,
51
+ "tokens/train_per_sec_per_gpu": 420.1,
52
+ "tokens/trainable": 3899782
53
+ },
54
+ {
55
+ "epoch": 0.8205128205128205,
56
+ "grad_norm": 2.4375,
57
+ "learning_rate": 7.75e-06,
58
+ "loss": 0.8140869140625,
59
+ "memory/device_reserved (GiB)": 89.53,
60
+ "memory/max_active (GiB)": 79.77,
61
+ "memory/max_allocated (GiB)": 79.77,
62
+ "ppl": 2.25711,
63
+ "step": 4,
64
+ "tokens/total": 8388608,
65
+ "tokens/train_per_sec_per_gpu": 446.85,
66
+ "tokens/trainable": 5212890
67
+ },
68
+ {
69
+ "epoch": 1.0,
70
+ "grad_norm": 2.0,
71
+ "learning_rate": 3.2500000000000015e-06,
72
+ "loss": 0.82373046875,
73
+ "memory/device_reserved (GiB)": 89.53,
74
+ "memory/max_active (GiB)": 79.77,
75
+ "memory/max_allocated (GiB)": 79.77,
76
+ "ppl": 2.27899,
77
+ "step": 5,
78
+ "tokens/total": 10223616,
79
+ "tokens/train_per_sec_per_gpu": 503.81,
80
+ "tokens/trainable": 6345970
81
+ }
82
+ ],
83
+ "logging_steps": 1,
84
+ "max_steps": 5,
85
+ "num_input_tokens_seen": 0,
86
+ "num_train_epochs": 1,
87
+ "save_steps": 5,
88
+ "stateful_callbacks": {
89
+ "TrainerControl": {
90
+ "args": {
91
+ "should_epoch_stop": false,
92
+ "should_evaluate": false,
93
+ "should_log": false,
94
+ "should_save": true,
95
+ "should_training_stop": true
96
+ },
97
+ "attributes": {}
98
+ }
99
+ },
100
+ "total_flos": 6.855380953947177e+17,
101
+ "train_batch_size": 8,
102
+ "trial_name": null,
103
+ "trial_params": null
104
+ }