Hugging Face's logo

llmflow
/
krx_Qwen2_7B_Instruct_v1_1

Text Generation
Transformers
Safetensors
English
qwen2
text-generation-inference
unsloth
trl
sft
krx
conversational
Inference Endpoints
Model card Files Community
llmflow commited on
Commit
bfd32cf
·
verified ·
1 Parent(s): 0bdd6ce

Trained with Unsloth

Browse files
Files changed (2) hide show
  1. config.json +5 -2
  2. generation_config.json +1 -1
config.json CHANGED
@@ -10,7 +10,7 @@
10
  "hidden_size": 3584,
11
  "initializer_range": 0.02,
12
  "intermediate_size": 18944,
13
- "max_position_embeddings": 32768,
14
  "max_window_layers": 28,
15
  "model_type": "qwen2",
16
  "num_attention_heads": 28,
@@ -18,7 +18,10 @@
18
  "num_key_value_heads": 4,
19
  "pad_token_id": 151643,
20
  "rms_norm_eps": 1e-06,
21
- "rope_scaling": null,
 
 
 
22
  "rope_theta": 1000000.0,
23
  "sliding_window": null,
24
  "tie_word_embeddings": false,
 
10
  "hidden_size": 3584,
11
  "initializer_range": 0.02,
12
  "intermediate_size": 18944,
13
+ "max_position_embeddings": 50000,
14
  "max_window_layers": 28,
15
  "model_type": "qwen2",
16
  "num_attention_heads": 28,
 
18
  "num_key_value_heads": 4,
19
  "pad_token_id": 151643,
20
  "rms_norm_eps": 1e-06,
21
+ "rope_scaling": {
22
+ "factor": 1.52587890625,
23
+ "type": "linear"
24
+ },
25
  "rope_theta": 1000000.0,
26
  "sliding_window": null,
27
  "tie_word_embeddings": false,
generation_config.json CHANGED
@@ -5,7 +5,7 @@
5
  151645,
6
  151643
7
  ],
8
- "max_length": 32768,
9
  "pad_token_id": 151643,
10
  "repetition_penalty": 1.05,
11
  "temperature": 0.7,
 
5
  151645,
6
  151643
7
  ],
8
+ "max_length": 50000,
9
  "pad_token_id": 151643,
10
  "repetition_penalty": 1.05,
11
  "temperature": 0.7,