Llama 3.3 8B 128K Instruct (Fixed)
Original allura-forge/Llama-3.3-8B-Instruct, Thanks!
Additional Fixes:
- Added
rope_scaling - Added chat template (Unsloth) in tokenizer config
- Updated generation config
- Enabled full context length
- Downloads last month
- 718
Model tree for shb777/Llama-3.3-8B-Instruct-128K
Base model
allura-forge/Llama-3.3-8B-InstructEvaluation results
- acc_norm on BBHself-reported54.100
- acc_norm on GPQAself-reported29.900
- acc on MMLU Proself-reported38.000
- acc_norm on MuSRself-reported37.800
- avg(prompt_strict + inst_strict) on IFEvalself-reported85.200
- exact_match on MATH Hardself-reported27.300