Create config.json
Browse files- config.json +1 -68
config.json
CHANGED
@@ -1,70 +1,3 @@
|
|
1 |
{
|
2 |
-
|
3 |
-
"Qwen2ForCausalLM"
|
4 |
-
],
|
5 |
-
"attention_dropout": 0.0,
|
6 |
-
"bos_token_id": 151643,
|
7 |
-
"eos_token_id": 151645,
|
8 |
-
"hidden_act": "silu",
|
9 |
-
"hidden_size": 5120,
|
10 |
-
"initializer_range": 0.02,
|
11 |
-
"intermediate_size": 13824,
|
12 |
-
"max_position_embeddings": 32768,
|
13 |
-
"max_window_layers": 48,
|
14 |
-
"model_type": "qwen2",
|
15 |
-
"num_attention_heads": 40,
|
16 |
-
"num_hidden_layers": 48,
|
17 |
-
"num_key_value_heads": 8,
|
18 |
-
"quantization_config": {
|
19 |
-
"config_groups": {
|
20 |
-
"group_0": {
|
21 |
-
"input_activations": {
|
22 |
-
"actorder": null,
|
23 |
-
"block_structure": null,
|
24 |
-
"dynamic": true,
|
25 |
-
"group_size": null,
|
26 |
-
"num_bits": 8,
|
27 |
-
"observer": null,
|
28 |
-
"observer_kwargs": {},
|
29 |
-
"strategy": "token",
|
30 |
-
"symmetric": true,
|
31 |
-
"type": "float"
|
32 |
-
},
|
33 |
-
"output_activations": null,
|
34 |
-
"targets": [
|
35 |
-
"Linear"
|
36 |
-
],
|
37 |
-
"weights": {
|
38 |
-
"actorder": null,
|
39 |
-
"block_structure": null,
|
40 |
-
"dynamic": false,
|
41 |
-
"group_size": null,
|
42 |
-
"num_bits": 8,
|
43 |
-
"observer": "minmax",
|
44 |
-
"observer_kwargs": {},
|
45 |
-
"strategy": "channel",
|
46 |
-
"symmetric": true,
|
47 |
-
"type": "float"
|
48 |
-
}
|
49 |
-
}
|
50 |
-
},
|
51 |
-
"format": "float-quantized",
|
52 |
-
"global_compression_ratio": 1.531585969025769,
|
53 |
-
"ignore": [
|
54 |
-
"lm_head"
|
55 |
-
],
|
56 |
-
"kv_cache_scheme": null,
|
57 |
-
"quant_method": "compressed-tensors",
|
58 |
-
"quantization_status": "compressed"
|
59 |
-
},
|
60 |
-
"rms_norm_eps": 1e-06,
|
61 |
-
"rope_scaling": null,
|
62 |
-
"rope_theta": 1000000.0,
|
63 |
-
"sliding_window": 131072,
|
64 |
-
"tie_word_embeddings": false,
|
65 |
-
"torch_dtype": "bfloat16",
|
66 |
-
"transformers_version": "4.50.1",
|
67 |
-
"use_cache": true,
|
68 |
-
"use_sliding_window": false,
|
69 |
-
"vocab_size": 152064
|
70 |
}
|
|
|
1 |
{
|
2 |
+
"model_type": "qwen2"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
3 |
}
|