Upload LlavaForConditionalGeneration
Browse files- config.json +61 -0
- generation_config.json +7 -0
- model.safetensors +3 -0
config.json
ADDED
|
@@ -0,0 +1,61 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "llava-1.5-7b-hf-awq",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"LlavaForConditionalGeneration"
|
| 5 |
+
],
|
| 6 |
+
"ignore_index": -100,
|
| 7 |
+
"image_token_index": 32000,
|
| 8 |
+
"model_type": "llava",
|
| 9 |
+
"pad_token_id": 32001,
|
| 10 |
+
"projector_hidden_act": "gelu",
|
| 11 |
+
"quantizaion_config": {
|
| 12 |
+
"backend": "autoawq",
|
| 13 |
+
"bits": 4,
|
| 14 |
+
"do_fuse": false,
|
| 15 |
+
"fuse_max_seq_len": null,
|
| 16 |
+
"group_size": 128,
|
| 17 |
+
"modules_to_fuse": null,
|
| 18 |
+
"modules_to_not_convert": null,
|
| 19 |
+
"quant_method": "awq",
|
| 20 |
+
"version": "gemm",
|
| 21 |
+
"zero_point": true
|
| 22 |
+
},
|
| 23 |
+
"quantization_config": {
|
| 24 |
+
"bits": 4,
|
| 25 |
+
"group_size": 128,
|
| 26 |
+
"modules_to_not_convert": [
|
| 27 |
+
"vision_tower",
|
| 28 |
+
"multi_modal_projector"
|
| 29 |
+
],
|
| 30 |
+
"quant_method": "awq",
|
| 31 |
+
"version": "gemm",
|
| 32 |
+
"zero_point": true
|
| 33 |
+
},
|
| 34 |
+
"text_config": {
|
| 35 |
+
"_name_or_path": "meta-llama/Llama-2-7b-hf",
|
| 36 |
+
"architectures": [
|
| 37 |
+
"LlamaForCausalLM"
|
| 38 |
+
],
|
| 39 |
+
"max_position_embeddings": 4096,
|
| 40 |
+
"model_type": "llama",
|
| 41 |
+
"rms_norm_eps": 1e-05,
|
| 42 |
+
"torch_dtype": "float16",
|
| 43 |
+
"vocab_size": 32064
|
| 44 |
+
},
|
| 45 |
+
"torch_dtype": "float16",
|
| 46 |
+
"transformers_version": "4.36.0.dev0",
|
| 47 |
+
"vision_config": {
|
| 48 |
+
"hidden_size": 1024,
|
| 49 |
+
"image_size": 336,
|
| 50 |
+
"intermediate_size": 4096,
|
| 51 |
+
"model_type": "clip_vision_model",
|
| 52 |
+
"num_attention_heads": 16,
|
| 53 |
+
"num_hidden_layers": 24,
|
| 54 |
+
"patch_size": 14,
|
| 55 |
+
"projection_dim": 768,
|
| 56 |
+
"vocab_size": 32000
|
| 57 |
+
},
|
| 58 |
+
"vision_feature_layer": -2,
|
| 59 |
+
"vision_feature_select_strategy": "default",
|
| 60 |
+
"vocab_size": 32064
|
| 61 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 1,
|
| 4 |
+
"eos_token_id": 2,
|
| 5 |
+
"pad_token_id": 32001,
|
| 6 |
+
"transformers_version": "4.36.0.dev0"
|
| 7 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cf17622b203bfc1dea9da031ee44728a417c571e92c1ab89153275c8d134824f
|
| 3 |
+
size 4539479520
|