MishaGGG commited on
Commit
4f3f41f
·
verified ·
1 Parent(s): df60b5d

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +19 -1
README.md CHANGED
@@ -1,4 +1,22 @@
1
  ---
2
  license: mit
3
  ---
4
- A simple AI tester. It supports HF and GGUF formats. Open `main.py`; you'll figure it out in 5 minutes.
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
  license: mit
3
  ---
4
+ A simple AI tester. It supports HF and GGUF formats. Open `main.py`; you'll figure it out in 5 minutes.
5
+
6
+ Dataset Path
7
+ DATASET_PATH = "gramm.jsonl" — path to your JSONL file containing questions and answers.
8
+
9
+ Model Path
10
+ MODEL_PATH = "model/" — folder with your local HuggingFace model or GGUF file.
11
+
12
+ Model Format Switch
13
+ USE_GGUF = False — set to True for GGUF models, False for HuggingFace Transformers.
14
+
15
+ Generation Temperature
16
+ TEMPERATURE = 0.6 — controls prediction randomness; lower values are more deterministic.
17
+
18
+ JSONL Format
19
+ Each line needs "q" for the question prompt and "a" for space-separated correct answers.
20
+
21
+ Multi-Answer Support
22
+ The "a" field accepts multiple answers separated by spaces, supporting questions with several correct options.