PyTorch
English
Tevatron
phi3_v
vidore
custom_code
MrLight commited on
Commit
c1385b6
1 Parent(s): b2e183b

Update config.json

Browse files
Files changed (1) hide show
  1. config.json +8 -10
config.json CHANGED
@@ -1,12 +1,12 @@
1
  {
2
- "_name_or_path": "MrLight/dse-phi3-v1.1-ft",
3
  "architectures": [
4
  "Phi3VForCausalLM"
5
  ],
6
  "attention_dropout": 0.0,
7
  "auto_map": {
8
- "AutoConfig": "microsoft/Phi-3-vision-128k-instruct--configuration_phi3_v.Phi3VConfig",
9
- "AutoModelForCausalLM": "microsoft/Phi-3-vision-128k-instruct--modeling_phi3_v.Phi3VForCausalLM"
10
  },
11
  "bos_token_id": 1,
12
  "embd_layer": {
@@ -16,7 +16,6 @@
16
  "use_hd_transform": true,
17
  "with_learnable_separator": true
18
  },
19
- "embd_pdrop": 0.0,
20
  "eos_token_id": 2,
21
  "hidden_act": "silu",
22
  "hidden_size": 3072,
@@ -34,8 +33,6 @@
34
  "num_hidden_layers": 32,
35
  "num_key_value_heads": 32,
36
  "original_max_position_embeddings": 4096,
37
- "pad_token_id": 32000,
38
- "resid_pdrop": 0.0,
39
  "rms_norm_eps": 1e-05,
40
  "rope_scaling": {
41
  "long_factor": [
@@ -144,7 +141,8 @@
144
  "sliding_window": 131072,
145
  "tie_word_embeddings": false,
146
  "torch_dtype": "bfloat16",
147
- "transformers_version": "4.41.2",
148
- "use_cache": false,
149
- "vocab_size": 32064
150
- }
 
 
1
  {
2
+ "_name_or_path": "dse-phi3-docmatix-v1.0",
3
  "architectures": [
4
  "Phi3VForCausalLM"
5
  ],
6
  "attention_dropout": 0.0,
7
  "auto_map": {
8
+ "AutoConfig": "configuration_phi3_v.Phi3VConfig",
9
+ "AutoModelForCausalLM": "modeling_phi3_v.Phi3VForCausalLM"
10
  },
11
  "bos_token_id": 1,
12
  "embd_layer": {
 
16
  "use_hd_transform": true,
17
  "with_learnable_separator": true
18
  },
 
19
  "eos_token_id": 2,
20
  "hidden_act": "silu",
21
  "hidden_size": 3072,
 
33
  "num_hidden_layers": 32,
34
  "num_key_value_heads": 32,
35
  "original_max_position_embeddings": 4096,
 
 
36
  "rms_norm_eps": 1e-05,
37
  "rope_scaling": {
38
  "long_factor": [
 
141
  "sliding_window": 131072,
142
  "tie_word_embeddings": false,
143
  "torch_dtype": "bfloat16",
144
+ "transformers_version": "4.38.1",
145
+ "use_cache": true,
146
+ "vocab_size": 32064,
147
+ "_attn_implementation": "flash_attention_2"
148
+ }