CyberNative
/

CyberBase-13b

Text Generation

text-generation-inference

8-bit precision

Model card Files Files and versions

CyberNative commited on Aug 18, 2023

Commit

618ddcb

·

1 Parent(s): 184c7b6

Update README.md

Files changed (1) hide show

README.md +1 -1

README.md CHANGED Viewed

@@ -11,7 +11,7 @@ CyberBase 13b 8k *base model* - (llama-2-13b - lmsys/vicuna-13b-v1.5-16k)
 - sequence_len: 4096
 - max_packed_sequence_len: 4096
 - lora_r: 256
-- lora_alpha: 256
 - trainable params: 1,001,390,080 || all params: 14,017,264,640 || trainable%: 7.143976415643959
 # Base cybersecurity model for future fine-tuning, it is not recomended to use on it's own.

 - sequence_len: 4096
 - max_packed_sequence_len: 4096
 - lora_r: 256
+- lora_alpha: 128
 - trainable params: 1,001,390,080 || all params: 14,017,264,640 || trainable%: 7.143976415643959
 # Base cybersecurity model for future fine-tuning, it is not recomended to use on it's own.