cstr commited on
Commit
c1ecce2
1 Parent(s): 9c5ef43

Upload folder using huggingface_hub

Browse files
Files changed (3) hide show
  1. config.json +49 -0
  2. generation_config.json +10 -0
  3. model.safetensors +3 -0
config.json ADDED
@@ -0,0 +1,49 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "cstr/wmt21-dense-24-wide-en-x-st",
3
+ "activation_dropout": 0.0,
4
+ "activation_function": "relu",
5
+ "architectures": [
6
+ "M2M100ForConditionalGeneration"
7
+ ],
8
+ "attention_dropout": 0.1,
9
+ "bos_token_id": 0,
10
+ "d_model": 2048,
11
+ "decoder_attention_heads": 32,
12
+ "decoder_ffn_dim": 16384,
13
+ "decoder_layerdrop": 0,
14
+ "decoder_layers": 24,
15
+ "decoder_start_token_id": 2,
16
+ "dropout": 0.1,
17
+ "encoder_attention_heads": 32,
18
+ "encoder_ffn_dim": 16384,
19
+ "encoder_layerdrop": 0,
20
+ "encoder_layers": 24,
21
+ "eos_token_id": 2,
22
+ "init_std": 0.02,
23
+ "is_encoder_decoder": true,
24
+ "max_length": 200,
25
+ "max_position_embeddings": 1024,
26
+ "model_type": "m2m_100",
27
+ "num_beams": 5,
28
+ "num_hidden_layers": 24,
29
+ "pad_token_id": 1,
30
+ "quantization_config": {
31
+ "_load_in_4bit": false,
32
+ "_load_in_8bit": true,
33
+ "bnb_4bit_compute_dtype": "float32",
34
+ "bnb_4bit_quant_type": "fp4",
35
+ "bnb_4bit_use_double_quant": false,
36
+ "llm_int8_enable_fp32_cpu_offload": false,
37
+ "llm_int8_has_fp16_weight": false,
38
+ "llm_int8_skip_modules": null,
39
+ "llm_int8_threshold": 6.0,
40
+ "load_in_4bit": false,
41
+ "load_in_8bit": true,
42
+ "quant_method": "bitsandbytes"
43
+ },
44
+ "scale_embedding": true,
45
+ "torch_dtype": "float16",
46
+ "transformers_version": "4.38.2",
47
+ "use_cache": true,
48
+ "vocab_size": 128009
49
+ }
generation_config.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 0,
4
+ "decoder_start_token_id": 2,
5
+ "eos_token_id": 2,
6
+ "max_length": 200,
7
+ "num_beams": 5,
8
+ "pad_token_id": 1,
9
+ "transformers_version": "4.38.2"
10
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f3e3719729741cb23ff7f9576eb6bd0a8da32e8f27fb9b501627e0beff3c8bad
3
+ size 4963515192