monsoon-nlp commited on
Commit
626ff8a
·
verified ·
1 Parent(s): d684024

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +15 -0
README.md CHANGED
@@ -35,6 +35,21 @@ in an exam format. **It is insufficient for training or finetuning an expert sys
35
  Patients need personalized, expert advice beyond what can be described on an exam
36
  or returned by an AI.
37
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
38
  ## Source information
39
 
40
  ### Source 1
 
35
  Patients need personalized, expert advice beyond what can be described on an exam
36
  or returned by an AI.
37
 
38
+ ### How to evaluate
39
+
40
+ You can run [this LightEval evaluation](https://github.com/mapmeld/lighteval-tasks/blob/main/community_tasks/gen_counselor_evals.py) with the command:
41
+
42
+ ```
43
+ lighteval accelerate \
44
+ "pretrained=meta-llama/Llama-3.1-8B-Instruct" \
45
+ "community|genetic-counselor-multiple-choice|0|0" \
46
+ --custom-tasks lighteval-tasks/community_tasks/gen_counselor_evals.py \
47
+ --override-batch-size=5 \
48
+ --use-chat-template
49
+ ```
50
+
51
+ Llama-3.1-8B-Instruct got 50.67% accuracy. I haven't been able to run all of the questions by OpenAI, but ChatGPT was doing pretty good on a sample of questions.
52
+
53
  ## Source information
54
 
55
  ### Source 1