Hugging Face's logo

weizhiwang
/
LLaVA-Llama-3-8B

Text Generation
Transformers
Safetensors
English
llava_llama
Inference Endpoints
Model card Files Community
5
Weizhi Wang commited on
Commit
2aff7fb
•
1 Parent(s): 3289f18

upgrade to CLIP-L-336px

Browse files
Files changed (7) hide show
  1. config.json +2 -2
  2. pytorch_model-00001-of-00002.bin +1 -1
  3. pytorch_model-00002-of-00002.bin +2 -2
  4. pytorch_model.bin.index.json +1 -1
  5. tokenizer_config.json +1 -1
  6. trainer_state.json +0 -0
  7. training_args.bin +1 -1
config.json CHANGED
@@ -21,7 +21,7 @@
21
  "mm_use_im_start_end": false,
22
  "mm_vision_select_feature": "patch",
23
  "mm_vision_select_layer": -2,
24
- "mm_vision_tower": "openai/clip-vit-large-patch14",
25
  "model_type": "llava",
26
  "num_attention_heads": 32,
27
  "num_hidden_layers": 32,
@@ -32,7 +32,7 @@
32
  "rope_scaling": null,
33
  "rope_theta": 500000.0,
34
  "tie_word_embeddings": false,
35
- "tokenizer_model_max_length": 4096,
36
  "tokenizer_padding_side": "right",
37
  "torch_dtype": "bfloat16",
38
  "transformers_version": "4.31.0",
 
21
  "mm_use_im_start_end": false,
22
  "mm_vision_select_feature": "patch",
23
  "mm_vision_select_layer": -2,
24
+ "mm_vision_tower": "openai/clip-vit-large-patch14-336",
25
  "model_type": "llava",
26
  "num_attention_heads": 32,
27
  "num_hidden_layers": 32,
 
32
  "rope_scaling": null,
33
  "rope_theta": 500000.0,
34
  "tie_word_embeddings": false,
35
+ "tokenizer_model_max_length": 2048,
36
  "tokenizer_padding_side": "right",
37
  "torch_dtype": "bfloat16",
38
  "transformers_version": "4.31.0",
pytorch_model-00001-of-00002.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:12f08708c94501aa2c3d04e6bb92f2c125b992792a1d377a472f73cf03867425
3
  size 9976553832
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:02396ce9183d20d5ae2893c65bc9d4611e667983bcacfa9761589f442f86d4a7
3
  size 9976553832
pytorch_model-00002-of-00002.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8cff318103cee2d722b8f01375ae74d740f6c5605e0b66097b320efb8479ff27
3
- size 6732552348
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:506bc30c144b2a87aa02d4b151e6577639371935690a03b1ed86f0cf25ea831a
3
+ size 6733207708
pytorch_model.bin.index.json CHANGED
@@ -1,6 +1,6 @@
1
  {
2
  "metadata": {
3
- "total_size": 16708845568
4
  },
5
  "weight_map": {
6
  "lm_head.weight": "pytorch_model-00002-of-00002.bin",
 
1
  {
2
  "metadata": {
3
+ "total_size": 16709500928
4
  },
5
  "weight_map": {
6
  "lm_head.weight": "pytorch_model-00002-of-00002.bin",
tokenizer_config.json CHANGED
@@ -2057,7 +2057,7 @@
2057
  "input_ids",
2058
  "attention_mask"
2059
  ],
2060
- "model_max_length": 4096,
2061
  "padding_side": "right",
2062
  "tokenizer_class": "PreTrainedTokenizerFast"
2063
  }
 
2057
  "input_ids",
2058
  "attention_mask"
2059
  ],
2060
+ "model_max_length": 2048,
2061
  "padding_side": "right",
2062
  "tokenizer_class": "PreTrainedTokenizerFast"
2063
  }
trainer_state.json ADDED
The diff for this file is too large to render. See raw diff
 
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2dec440fcf94b23f2fc3c9a99937f5b292eecc8df558b9b7cddf87dc70009654
3
  size 5947
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:408ff2c7ee46e11209df0570466e9f39961a21e99381323a300a9173b6fea03a
3
  size 5947