red pajama result

mistral results

comment out sleep safety (im lazy, i want it to go faster)

ph models

llama2 stuff

llama 7b (v1)

facebook old models result

gptj result

bloom results

updated compression / decompression scripts

all RWKV based results

initial model results?

result jsonl compression script

reset

Add lm-eval output files for arc_challenge

Add Git attributes and ignore files for JSONL and log files

initial commit