TroyDoesAI
commited on
Commit
•
484814c
1
Parent(s):
5ac3c5a
Need to remove this for merge
Browse files- config.json +34 -40
config.json
CHANGED
@@ -1,40 +1,34 @@
|
|
1 |
-
{
|
2 |
-
"_name_or_path": ".\\BlackSheep-Llama-3.2-3B",
|
3 |
-
"architectures": [
|
4 |
-
"LlamaForCausalLM"
|
5 |
-
],
|
6 |
-
"attention_bias": false,
|
7 |
-
"attention_dropout": 0.0,
|
8 |
-
"bos_token_id": 128000,
|
9 |
-
"eos_token_id": [
|
10 |
-
128001,
|
11 |
-
128008,
|
12 |
-
128009
|
13 |
-
],
|
14 |
-
"head_dim": 128,
|
15 |
-
"hidden_act": "silu",
|
16 |
-
"hidden_size": 3072,
|
17 |
-
"initializer_range": 0.02,
|
18 |
-
"intermediate_size": 8192,
|
19 |
-
"max_position_embeddings": 131072,
|
20 |
-
"mlp_bias": false,
|
21 |
-
"model_type": "llama",
|
22 |
-
"num_attention_heads": 24,
|
23 |
-
"num_hidden_layers": 28,
|
24 |
-
"num_key_value_heads": 8,
|
25 |
-
"pretraining_tp": 1,
|
26 |
-
"rms_norm_eps": 1e-05,
|
27 |
-
"rope_scaling":
|
28 |
-
|
29 |
-
|
30 |
-
|
31 |
-
|
32 |
-
|
33 |
-
|
34 |
-
|
35 |
-
"tie_word_embeddings": true,
|
36 |
-
"torch_dtype": "bfloat16",
|
37 |
-
"transformers_version": "4.44.2",
|
38 |
-
"use_cache": true,
|
39 |
-
"vocab_size": 128256
|
40 |
-
}
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": ".\\BlackSheep-Llama-3.2-3B",
|
3 |
+
"architectures": [
|
4 |
+
"LlamaForCausalLM"
|
5 |
+
],
|
6 |
+
"attention_bias": false,
|
7 |
+
"attention_dropout": 0.0,
|
8 |
+
"bos_token_id": 128000,
|
9 |
+
"eos_token_id": [
|
10 |
+
128001,
|
11 |
+
128008,
|
12 |
+
128009
|
13 |
+
],
|
14 |
+
"head_dim": 128,
|
15 |
+
"hidden_act": "silu",
|
16 |
+
"hidden_size": 3072,
|
17 |
+
"initializer_range": 0.02,
|
18 |
+
"intermediate_size": 8192,
|
19 |
+
"max_position_embeddings": 131072,
|
20 |
+
"mlp_bias": false,
|
21 |
+
"model_type": "llama",
|
22 |
+
"num_attention_heads": 24,
|
23 |
+
"num_hidden_layers": 28,
|
24 |
+
"num_key_value_heads": 8,
|
25 |
+
"pretraining_tp": 1,
|
26 |
+
"rms_norm_eps": 1e-05,
|
27 |
+
"rope_scaling": null
|
28 |
+
"rope_theta": 500000.0,
|
29 |
+
"tie_word_embeddings": true,
|
30 |
+
"torch_dtype": "bfloat16",
|
31 |
+
"transformers_version": "4.44.2",
|
32 |
+
"use_cache": true,
|
33 |
+
"vocab_size": 128256
|
34 |
+
}
|
|
|
|
|
|
|
|
|
|
|
|