dkleine commited on
Commit
1c81bd2
1 Parent(s): c80d904

Update README.md

Browse files

* changed to 4k context length (Phi-3-mini-4k-instruct)

Files changed (1) hide show
  1. README.md +1 -1
README.md CHANGED
@@ -221,7 +221,7 @@ Developers should apply responsible AI best practices and are responsible for en
221
 
222
  * Architecture: Phi-3 Mini has 3.8B parameters and is a dense decoder-only Transformer model. The model is fine-tuned with Supervised fine-tuning (SFT) and Direct Preference Optimization (DPO) to ensure alignment with human preferences and safety guidlines.
223
  * Inputs: Text. It is best suited for prompts using chat format.
224
- * Context length: 128K tokens
225
  * GPUS: 512 H100-80G
226
  * Training time: 7 days
227
  * Training data: 3.3T tokens
 
221
 
222
  * Architecture: Phi-3 Mini has 3.8B parameters and is a dense decoder-only Transformer model. The model is fine-tuned with Supervised fine-tuning (SFT) and Direct Preference Optimization (DPO) to ensure alignment with human preferences and safety guidlines.
223
  * Inputs: Text. It is best suited for prompts using chat format.
224
+ * Context length: 4K tokens
225
  * GPUS: 512 H100-80G
226
  * Training time: 7 days
227
  * Training data: 3.3T tokens