Text Generation
Transformers
Safetensors
starcoder2
code
Eval Results
text-generation-inference
Inference Endpoints
compressed-tensors
mgoin commited on
Commit
eda6098
1 Parent(s): b0c188d

Updated compression_config to quantization_config

Browse files
Files changed (1) hide show
  1. config.json +24 -24
config.json CHANGED
@@ -5,7 +5,29 @@
5
  ],
6
  "attention_dropout": 0.1,
7
  "bos_token_id": 0,
8
- "compression_config": {
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
9
  "config_groups": {
10
  "group_0": {
11
  "input_activations": null,
@@ -40,27 +62,5 @@
40
  "registry_requires_subclass": false,
41
  "sparsity_structure": "unstructured"
42
  }
43
- },
44
- "embedding_dropout": 0.1,
45
- "eos_token_id": 0,
46
- "hidden_act": "gelu_pytorch_tanh",
47
- "hidden_size": 3072,
48
- "initializer_range": 0.018042,
49
- "intermediate_size": 12288,
50
- "max_position_embeddings": 16384,
51
- "mlp_type": "default",
52
- "model_type": "starcoder2",
53
- "norm_epsilon": 1e-05,
54
- "norm_type": "layer_norm",
55
- "num_attention_heads": 24,
56
- "num_hidden_layers": 30,
57
- "num_key_value_heads": 2,
58
- "residual_dropout": 0.1,
59
- "rope_theta": 999999.4420358813,
60
- "sliding_window": 4096,
61
- "torch_dtype": "float32",
62
- "transformers_version": "4.43.3",
63
- "use_bias": true,
64
- "use_cache": true,
65
- "vocab_size": 49152
66
  }
 
5
  ],
6
  "attention_dropout": 0.1,
7
  "bos_token_id": 0,
8
+ "embedding_dropout": 0.1,
9
+ "eos_token_id": 0,
10
+ "hidden_act": "gelu_pytorch_tanh",
11
+ "hidden_size": 3072,
12
+ "initializer_range": 0.018042,
13
+ "intermediate_size": 12288,
14
+ "max_position_embeddings": 16384,
15
+ "mlp_type": "default",
16
+ "model_type": "starcoder2",
17
+ "norm_epsilon": 1e-05,
18
+ "norm_type": "layer_norm",
19
+ "num_attention_heads": 24,
20
+ "num_hidden_layers": 30,
21
+ "num_key_value_heads": 2,
22
+ "residual_dropout": 0.1,
23
+ "rope_theta": 999999.4420358813,
24
+ "sliding_window": 4096,
25
+ "torch_dtype": "float32",
26
+ "transformers_version": "4.43.3",
27
+ "use_bias": true,
28
+ "use_cache": true,
29
+ "vocab_size": 49152,
30
+ "quantization_config": {
31
  "config_groups": {
32
  "group_0": {
33
  "input_activations": null,
 
62
  "registry_requires_subclass": false,
63
  "sparsity_structure": "unstructured"
64
  }
65
+ }
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
66
  }