thanks to SakanaAI ❤
Browse files- config.json +78 -0
- generation_config.json +10 -0
- model-00001-of-00002.safetensors +3 -0
- model-00002-of-00002.safetensors +3 -0
- model.safetensors.index.json +0 -0
config.json
ADDED
@@ -0,0 +1,78 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "SakanaAI/EvoVLM-JP-v1-7B",
|
3 |
+
"architectures": [
|
4 |
+
"LlavaForConditionalGeneration"
|
5 |
+
],
|
6 |
+
"ignore_index": -100,
|
7 |
+
"image_grid_pinpoints": [
|
8 |
+
[
|
9 |
+
336,
|
10 |
+
672
|
11 |
+
],
|
12 |
+
[
|
13 |
+
672,
|
14 |
+
336
|
15 |
+
],
|
16 |
+
[
|
17 |
+
672,
|
18 |
+
672
|
19 |
+
],
|
20 |
+
[
|
21 |
+
1008,
|
22 |
+
336
|
23 |
+
],
|
24 |
+
[
|
25 |
+
336,
|
26 |
+
1008
|
27 |
+
]
|
28 |
+
],
|
29 |
+
"image_token_index": 32000,
|
30 |
+
"model_type": "llava",
|
31 |
+
"pad_token_id": 32001,
|
32 |
+
"projector_hidden_act": "gelu",
|
33 |
+
"quantization_config": {
|
34 |
+
"_load_in_4bit": false,
|
35 |
+
"_load_in_8bit": true,
|
36 |
+
"bnb_4bit_compute_dtype": "float32",
|
37 |
+
"bnb_4bit_quant_type": "fp4",
|
38 |
+
"bnb_4bit_use_double_quant": false,
|
39 |
+
"llm_int8_enable_fp32_cpu_offload": false,
|
40 |
+
"llm_int8_has_fp16_weight": false,
|
41 |
+
"llm_int8_skip_modules": null,
|
42 |
+
"llm_int8_threshold": 6.0,
|
43 |
+
"load_in_4bit": false,
|
44 |
+
"load_in_8bit": true,
|
45 |
+
"quant_method": "bitsandbytes"
|
46 |
+
},
|
47 |
+
"text_config": {
|
48 |
+
"_name_or_path": "llava_v1.6_mistral_7b-shisa",
|
49 |
+
"architectures": [
|
50 |
+
"MistralForCausalLM"
|
51 |
+
],
|
52 |
+
"intermediate_size": 14336,
|
53 |
+
"max_position_embeddings": 32768,
|
54 |
+
"model_type": "mistral",
|
55 |
+
"num_key_value_heads": 8,
|
56 |
+
"rms_norm_eps": 1e-05,
|
57 |
+
"sliding_window": 4096,
|
58 |
+
"torch_dtype": "float16",
|
59 |
+
"vocab_size": 32064
|
60 |
+
},
|
61 |
+
"torch_dtype": "float16",
|
62 |
+
"transformers_version": "4.38.2",
|
63 |
+
"use_image_newline_parameter": true,
|
64 |
+
"vision_config": {
|
65 |
+
"hidden_size": 1024,
|
66 |
+
"image_size": 336,
|
67 |
+
"intermediate_size": 4096,
|
68 |
+
"model_type": "clip_vision_model",
|
69 |
+
"num_attention_heads": 16,
|
70 |
+
"num_hidden_layers": 24,
|
71 |
+
"patch_size": 14,
|
72 |
+
"projection_dim": 768,
|
73 |
+
"vocab_size": 32000
|
74 |
+
},
|
75 |
+
"vision_feature_layer": -2,
|
76 |
+
"vision_feature_select_strategy": "default",
|
77 |
+
"vocab_size": 32064
|
78 |
+
}
|
generation_config.json
ADDED
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_from_model_config": true,
|
3 |
+
"bos_token_id": 1,
|
4 |
+
"eos_token_id": 2,
|
5 |
+
"max_new_tokens": 1024,
|
6 |
+
"num_beams": 5,
|
7 |
+
"pad_token_id": 32001,
|
8 |
+
"repetition_penalty": 1.5,
|
9 |
+
"transformers_version": "4.38.2"
|
10 |
+
}
|
model-00001-of-00002.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:00e89ecc1208949b71a567902427972a1107784637e05e315036bf1747b58067
|
3 |
+
size 4997539712
|
model-00002-of-00002.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:6e65fb6f9dc82a243b19c3b4cd899b06575b55f7339e7f57ac196eba015207be
|
3 |
+
size 2840222096
|
model.safetensors.index.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|