Seonghyeon Lee
sh0416
AI & ML interests
None yet
Organizations
sh0416's activity
Adding new benchmark dataset
#84 opened about 2 months ago
by
sh0416
Question about the correctness of solution
2
#8 opened about 2 months ago
by
sh0416
Question about the results for codellama/CodeLlama-7b-Instruct-hf
2
#20 opened about 1 year ago
by
sh0416
Question about their name.. why it is 2b???
2
#36 opened 8 months ago
by
sh0416
Request on prompts for the evaluation results.
#5 opened 8 months ago
by
sh0416
Add architectures field to config.json
1
#2 opened 8 months ago
by
chargoddard
check5 and check6 validity
#6 opened 9 months ago
by
sh0416
Supported programming language and their format.
#5 opened 9 months ago
by
sh0416
Could I get a simple workflow to add custom tokenizer?
1
#3 opened 10 months ago
by
sh0416
How to reproduce humaneval-python performance of codellama/CodeLlama-34b-Instruct-hf on the leaderboard?
2
#15 opened about 1 year ago
by
sanqiang
Reproduce pass@10 (temperature 0.8, top p 0.95)
#18 opened about 1 year ago
by
sh0416
Precise definition of personality in this dataset
1
#2 opened about 1 year ago
by
sh0416
Question about model architecture
1
#40 opened about 1 year ago
by
sh0416
Tokenization issue
1
#2 opened about 1 year ago
by
sh0416
Reproduce humaneval pass@10 and pass@100
2
#38 opened over 1 year ago
by
sh0416
Missing problem in humaneval
2
#1 opened over 1 year ago
by
sh0416
How to reproduce Table 7?
#36 opened over 1 year ago
by
sh0416