mervinpraison commited on
Commit
af40fab
1 Parent(s): 5e3c3de

Upload IdeficsForVisionText2Text

Browse files
Files changed (3) hide show
  1. config.json +63 -0
  2. generation_config.json +7 -0
  3. model.safetensors +3 -0
config.json ADDED
@@ -0,0 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "HuggingFaceM4/idefics-9b",
3
+ "additional_vocab_size": 2,
4
+ "alpha_initializer": "zeros",
5
+ "alpha_type": "float",
6
+ "alphas_initializer_range": 0.0,
7
+ "architectures": [
8
+ "IdeficsForVisionText2Text"
9
+ ],
10
+ "bos_token_id": 1,
11
+ "cross_layer_activation_function": "swiglu",
12
+ "cross_layer_interval": 4,
13
+ "dropout": 0.0,
14
+ "eos_token_id": 2,
15
+ "freeze_lm_head": false,
16
+ "freeze_text_layers": false,
17
+ "freeze_text_module_exceptions": [],
18
+ "freeze_vision_layers": false,
19
+ "freeze_vision_module_exceptions": [],
20
+ "hidden_act": "silu",
21
+ "hidden_size": 4096,
22
+ "initializer_range": 0.02,
23
+ "intermediate_size": 11008,
24
+ "max_position_embeddings": 2048,
25
+ "max_sequence_length": 2048,
26
+ "model_type": "idefics",
27
+ "num_attention_heads": 32,
28
+ "num_hidden_layers": 32,
29
+ "pad_token_id": 0,
30
+ "perceiver_config": {
31
+ "model_type": "idefics",
32
+ "qk_layer_norms_perceiver": true
33
+ },
34
+ "qk_layer_norms": true,
35
+ "quantization_config": {
36
+ "_load_in_4bit": true,
37
+ "_load_in_8bit": false,
38
+ "bnb_4bit_compute_dtype": "float16",
39
+ "bnb_4bit_quant_type": "nf4",
40
+ "bnb_4bit_use_double_quant": true,
41
+ "llm_int8_enable_fp32_cpu_offload": false,
42
+ "llm_int8_has_fp16_weight": false,
43
+ "llm_int8_skip_modules": [
44
+ "lm_head",
45
+ "embed_tokens"
46
+ ],
47
+ "llm_int8_threshold": 6.0,
48
+ "load_in_4bit": true,
49
+ "load_in_8bit": false,
50
+ "quant_method": "bitsandbytes"
51
+ },
52
+ "rms_norm_eps": 1e-06,
53
+ "tie_word_embeddings": false,
54
+ "torch_dtype": "float16",
55
+ "transformers_version": "4.38.2",
56
+ "use_cache": true,
57
+ "use_resampler": true,
58
+ "vision_config": {
59
+ "embed_dim": 1280,
60
+ "model_type": "idefics"
61
+ },
62
+ "vocab_size": 32000
63
+ }
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": 0,
6
+ "transformers_version": "4.38.2"
7
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:35a6077aed929213e2bf7ac72ec0b5a74a2cd493d96fd3c339112f80b1db48f1
3
+ size 4999822634