danielhanchen
commited on
Commit
•
df58268
1
Parent(s):
e44f61a
Upload Qwen2ForCausalLM
Browse files- config.json +4 -1
- generation_config.json +3 -2
config.json
CHANGED
@@ -16,12 +16,15 @@
|
|
16 |
"num_attention_heads": 14,
|
17 |
"num_hidden_layers": 24,
|
18 |
"num_key_value_heads": 2,
|
|
|
19 |
"rms_norm_eps": 1e-06,
|
|
|
20 |
"rope_theta": 1000000.0,
|
21 |
"sliding_window": null,
|
22 |
"tie_word_embeddings": true,
|
23 |
"torch_dtype": "bfloat16",
|
24 |
-
"transformers_version": "4.
|
|
|
25 |
"use_cache": true,
|
26 |
"use_sliding_window": false,
|
27 |
"vocab_size": 151936
|
|
|
16 |
"num_attention_heads": 14,
|
17 |
"num_hidden_layers": 24,
|
18 |
"num_key_value_heads": 2,
|
19 |
+
"pad_token_id": 151665,
|
20 |
"rms_norm_eps": 1e-06,
|
21 |
+
"rope_scaling": null,
|
22 |
"rope_theta": 1000000.0,
|
23 |
"sliding_window": null,
|
24 |
"tie_word_embeddings": true,
|
25 |
"torch_dtype": "bfloat16",
|
26 |
+
"transformers_version": "4.46.1",
|
27 |
+
"unsloth_fixed": true,
|
28 |
"use_cache": true,
|
29 |
"use_sliding_window": false,
|
30 |
"vocab_size": 151936
|
generation_config.json
CHANGED
@@ -5,10 +5,11 @@
|
|
5 |
151645,
|
6 |
151643
|
7 |
],
|
8 |
-
"
|
|
|
9 |
"repetition_penalty": 1.1,
|
10 |
"temperature": 0.7,
|
11 |
"top_k": 20,
|
12 |
"top_p": 0.8,
|
13 |
-
"transformers_version": "4.
|
14 |
}
|
|
|
5 |
151645,
|
6 |
151643
|
7 |
],
|
8 |
+
"max_length": 32768,
|
9 |
+
"pad_token_id": 151665,
|
10 |
"repetition_penalty": 1.1,
|
11 |
"temperature": 0.7,
|
12 |
"top_k": 20,
|
13 |
"top_p": 0.8,
|
14 |
+
"transformers_version": "4.46.1"
|
15 |
}
|