woodchen7 commited on
Commit
14bb5a4
·
verified ·
1 Parent(s): 9c529d9

Upload config.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. config.json +64 -0
config.json ADDED
@@ -0,0 +1,64 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "SeedOssForCausalLM"
4
+ ],
5
+ "attention_bias": true,
6
+ "attention_dropout": 0.1,
7
+ "attention_out_bias": false,
8
+ "bos_token_id": 0,
9
+ "eos_token_id": 2,
10
+ "head_dim": 128,
11
+ "hidden_act": "silu",
12
+ "hidden_size": 5120,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 27648,
15
+ "max_position_embeddings": 524288,
16
+ "mlp_bias": false,
17
+ "model_type": "seed_oss",
18
+ "num_attention_heads": 80,
19
+ "num_hidden_layers": 64,
20
+ "num_key_value_heads": 8,
21
+ "pad_token_id": 1,
22
+ "pretraining_tp": 1,
23
+ "quantization_config": {
24
+ "config_groups": {
25
+ "group_0": {
26
+ "input_activations": {
27
+ "dynamic": false,
28
+ "num_bits": 8,
29
+ "strategy": "tensor",
30
+ "type": "float"
31
+ },
32
+ "output_activations": null,
33
+ "targets": [
34
+ "Linear"
35
+ ],
36
+ "weights": {
37
+ "dynamic": false,
38
+ "num_bits": 8,
39
+ "strategy": "tensor",
40
+ "type": "float"
41
+ }
42
+ }
43
+ },
44
+ "format": "naive-quantized",
45
+ "ignored_layers": [
46
+ "lm_head",
47
+ "model.embed_tokens"
48
+ ],
49
+ "kv_cache_scheme": null,
50
+ "quant_method": "compressed-tensors",
51
+ "quantization_status": "compressed"
52
+ },
53
+ "residual_dropout": 0.1,
54
+ "rms_norm_eps": 1e-06,
55
+ "rope_scaling": {
56
+ "rope_type": "default"
57
+ },
58
+ "rope_theta": 10000000.0,
59
+ "tie_word_embeddings": false,
60
+ "torch_dtype": "bfloat16",
61
+ "transformers_version": "4.56.0.dev0",
62
+ "use_cache": false,
63
+ "vocab_size": 155136
64
+ }