danielhanchen commited on
Commit
64141e7
·
verified ·
1 Parent(s): ed389b0

Add files using upload-large-folder tool

Browse files
Files changed (3) hide show
  1. config.json +29 -1
  2. tokenizer.json +2 -2
  3. tokenizer_config.json +4 -3
config.json CHANGED
@@ -3,32 +3,60 @@
3
  "Gemma3ForConditionalGeneration"
4
  ],
5
  "boi_token_index": 255999,
 
6
  "eoi_token_index": 256000,
 
7
  "image_token_index": 262144,
8
  "initializer_range": 0.02,
9
  "mm_tokens_per_image": 256,
10
  "model_type": "gemma3",
 
11
  "text_config": {
 
 
 
 
 
 
 
12
  "hidden_size": 2560,
 
13
  "intermediate_size": 10240,
 
14
  "model_type": "gemma3_text",
 
15
  "num_hidden_layers": 34,
 
 
 
 
16
  "rope_scaling": {
17
  "factor": 8.0,
18
  "rope_type": "linear"
19
  },
20
- "sliding_window": 1024
 
 
 
 
 
21
  },
22
  "torch_dtype": "bfloat16",
23
  "transformers_version": "4.50.0.dev0",
 
24
  "vision_config": {
 
 
25
  "hidden_size": 1152,
26
  "image_size": 896,
27
  "intermediate_size": 4304,
 
28
  "model_type": "siglip_vision_model",
29
  "num_attention_heads": 16,
 
30
  "num_hidden_layers": 27,
31
  "patch_size": 14,
 
32
  "vision_use_head": false
33
  }
34
  }
 
3
  "Gemma3ForConditionalGeneration"
4
  ],
5
  "boi_token_index": 255999,
6
+ "bos_token_id": 2,
7
  "eoi_token_index": 256000,
8
+ "eos_token_id": 1,
9
  "image_token_index": 262144,
10
  "initializer_range": 0.02,
11
  "mm_tokens_per_image": 256,
12
  "model_type": "gemma3",
13
+ "pad_token_id": 0,
14
  "text_config": {
15
+ "attention_bias": false,
16
+ "attention_dropout": 0.0,
17
+ "attn_logit_softcapping": null,
18
+ "cache_implementation": "hybrid",
19
+ "final_logit_softcapping": null,
20
+ "head_dim": 256,
21
+ "hidden_activation": "gelu_pytorch_tanh",
22
  "hidden_size": 2560,
23
+ "initializer_range": 0.02,
24
  "intermediate_size": 10240,
25
+ "max_position_embeddings": 131072,
26
  "model_type": "gemma3_text",
27
+ "num_attention_heads": 8,
28
  "num_hidden_layers": 34,
29
+ "num_key_value_heads": 4,
30
+ "query_pre_attn_scalar": 256,
31
+ "rms_norm_eps": 1e-06,
32
+ "rope_local_base_freq": 10000.0,
33
  "rope_scaling": {
34
  "factor": 8.0,
35
  "rope_type": "linear"
36
  },
37
+ "rope_theta": 1000000.0,
38
+ "sliding_window": 1024,
39
+ "sliding_window_pattern": 6,
40
+ "torch_dtype": "bfloat16",
41
+ "use_cache": true,
42
+ "vocab_size": 262208
43
  },
44
  "torch_dtype": "bfloat16",
45
  "transformers_version": "4.50.0.dev0",
46
+ "unsloth_fixed": true,
47
  "vision_config": {
48
+ "attention_dropout": 0.0,
49
+ "hidden_act": "gelu_pytorch_tanh",
50
  "hidden_size": 1152,
51
  "image_size": 896,
52
  "intermediate_size": 4304,
53
+ "layer_norm_eps": 1e-06,
54
  "model_type": "siglip_vision_model",
55
  "num_attention_heads": 16,
56
+ "num_channels": 3,
57
  "num_hidden_layers": 27,
58
  "patch_size": 14,
59
+ "torch_dtype": "bfloat16",
60
  "vision_use_head": false
61
  }
62
  }
tokenizer.json CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7d4046bf0505a327dd5a0abbb427ecd4fc82f99c2ceaa170bc61ecde12809b0c
3
- size 33384570
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4667f2089529e8e7657cfb6d1c19910ae71ff5f28aa7ab2ff2763330affad795
3
+ size 33384568
tokenizer_config.json CHANGED
@@ -2160,7 +2160,7 @@
2160
  "normalized": false,
2161
  "rstrip": false,
2162
  "single_word": false,
2163
- "special": false
2164
  },
2165
  "256000": {
2166
  "content": "<end_of_image>",
@@ -2168,7 +2168,7 @@
2168
  "normalized": false,
2169
  "rstrip": false,
2170
  "single_word": false,
2171
- "special": false
2172
  },
2173
  "256001": {
2174
  "content": "<unused99>",
@@ -51334,8 +51334,9 @@
51334
  "image_token": "<image_soft_token>"
51335
  },
51336
  "image_token": "<image_soft_token>",
51337
- "model_max_length": 1000000000000000019884624838656,
51338
  "pad_token": "<pad>",
 
51339
  "processor_class": "Gemma3Processor",
51340
  "sp_model_kwargs": null,
51341
  "spaces_between_special_tokens": false,
 
2160
  "normalized": false,
2161
  "rstrip": false,
2162
  "single_word": false,
2163
+ "special": true
2164
  },
2165
  "256000": {
2166
  "content": "<end_of_image>",
 
2168
  "normalized": false,
2169
  "rstrip": false,
2170
  "single_word": false,
2171
+ "special": true
2172
  },
2173
  "256001": {
2174
  "content": "<unused99>",
 
51334
  "image_token": "<image_soft_token>"
51335
  },
51336
  "image_token": "<image_soft_token>",
51337
+ "model_max_length": 131072,
51338
  "pad_token": "<pad>",
51339
+ "padding_side": "left",
51340
  "processor_class": "Gemma3Processor",
51341
  "sp_model_kwargs": null,
51342
  "spaces_between_special_tokens": false,