Upload config.json
Browse files- config.json +29 -1
config.json
CHANGED
@@ -1,6 +1,7 @@
|
|
1 |
{
|
2 |
"alpha_pattern": {},
|
3 |
"auto_mapping": null,
|
|
|
4 |
"base_model_name_or_path": "abacusai/Llama-3-Smaug-8B",
|
5 |
"bias": "none",
|
6 |
"fan_in_fan_out": false,
|
@@ -28,5 +29,32 @@
|
|
28 |
],
|
29 |
"task_type": "SEQ_CLS",
|
30 |
"use_dora": false,
|
31 |
-
"use_rslora": false
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
32 |
}
|
|
|
1 |
{
|
2 |
"alpha_pattern": {},
|
3 |
"auto_mapping": null,
|
4 |
+
|
5 |
"base_model_name_or_path": "abacusai/Llama-3-Smaug-8B",
|
6 |
"bias": "none",
|
7 |
"fan_in_fan_out": false,
|
|
|
29 |
],
|
30 |
"task_type": "SEQ_CLS",
|
31 |
"use_dora": false,
|
32 |
+
"use_rslora": false,
|
33 |
+
"architectures": [
|
34 |
+
"LlamaForSequenceClassification"
|
35 |
+
],
|
36 |
+
"attention_bias": false,
|
37 |
+
"attention_dropout": 0.0,
|
38 |
+
"bos_token_id": 128000,
|
39 |
+
"eos_token_id": 128001,
|
40 |
+
"hidden_act": "silu",
|
41 |
+
"hidden_size": 4096,
|
42 |
+
"initializer_range": 0.02,
|
43 |
+
"intermediate_size": 14336,
|
44 |
+
"max_position_embeddings": 8192,
|
45 |
+
"model_type": "llama",
|
46 |
+
"num_attention_heads": 32,
|
47 |
+
"num_hidden_layers": 32,
|
48 |
+
"num_key_value_heads": 8,
|
49 |
+
"pretraining_tp": 1,
|
50 |
+
"rms_norm_eps": 1e-05,
|
51 |
+
"rope_scaling": null,
|
52 |
+
"rope_theta": 500000.0,
|
53 |
+
"tie_word_embeddings": false,
|
54 |
+
"torch_dtype": "bfloat16",
|
55 |
+
"transformers_version": "4.37.2",
|
56 |
+
"use_cache": true,
|
57 |
+
"vocab_size": 128256,
|
58 |
+
"num_labels": 1,
|
59 |
+
"problem_type": "regression"
|
60 |
}
|