Visual Question Answering
Transformers
English
videollama2_mistral
text-generation
multimodal large language model
large video-language model
Inference Endpoints
ClownRat commited on
Commit
bd38fd6
1 Parent(s): 634b54d

Model upload.

Browse files
Files changed (2) hide show
  1. config.json +41 -0
  2. mm_projector.bin +3 -0
config.json ADDED
@@ -0,0 +1,41 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "mistralai/Mistral-7B-Instruct-v0.2",
3
+ "architectures": [
4
+ "MistralForCausalLM"
5
+ ],
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 1,
8
+ "eos_token_id": 2,
9
+ "freeze_mm_mlp_adapter": false,
10
+ "hidden_act": "silu",
11
+ "hidden_size": 4096,
12
+ "image_aspect_ratio": "square",
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 14336,
15
+ "max_position_embeddings": 32768,
16
+ "mm_hidden_size": 1024,
17
+ "mm_projector_lr": null,
18
+ "mm_projector_type": "stc_connector",
19
+ "mm_use_im_patch_token": false,
20
+ "mm_use_im_start_end": false,
21
+ "mm_vision_select_feature": "patch",
22
+ "mm_vision_select_layer": -2,
23
+ "mm_vision_tower": "openai/clip-vit-large-patch14-336",
24
+ "model_type": "mistral",
25
+ "num_attention_heads": 32,
26
+ "num_frames": 16,
27
+ "num_hidden_layers": 32,
28
+ "num_key_value_heads": 8,
29
+ "rms_norm_eps": 1e-05,
30
+ "rope_theta": 1000000.0,
31
+ "sliding_window": null,
32
+ "tie_word_embeddings": false,
33
+ "tokenizer_model_max_length": 2048,
34
+ "tokenizer_padding_side": "right",
35
+ "torch_dtype": "bfloat16",
36
+ "transformers_version": "4.37.2",
37
+ "tune_mm_mlp_adapter": true,
38
+ "use_cache": false,
39
+ "use_mm_proj": true,
40
+ "vocab_size": 32000
41
+ }
mm_projector.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d49e42b8a43ee308d2856ebe0b7d7fd9aa7fe7b46426d072b22fef529ef1b02f
3
+ size 978413913