← All skills
evaluating-llms-harness
65 / 100 · Review before installing
Check before installing.
By artubss/SKILLS-CLAUDE-CODE (GitHub) · other · 11 stars · scanned at commit ac67ed8
Avalia LLMs em mais de 60 benchmarks acadêmicos (MMLU, HumanEval, GSM8K, TruthfulQA, HellaSwag). Use quando benchmarking de qualidade de modelo, comparação de modelos, relatório de resultados acadêmicos ou rastreamento de progresso de treinamento. Padrão da indústria usado por EleutherAI, HuggingFace e laboratórios principais. Suporta HuggingFace, vLLM, APIs.
Works with any agent that supports SKILL.md skills.
What this skill can do
Downloads and installs other software
The author does not mention this.
Found in 3 places.
Runs commands on your computer
The author does not mention this.
Found in 27 places.
Show 17 more
- skills/skills/ai-research/evaluation-lm-evaluation-harness/SKILL.md, line 67
--tasks mmlu,gsm8k,hellaswag,truthfulqa,arc_challenge - skills/skills/ai-research/evaluation-lm-evaluation-harness/SKILL.md, line 74
lm_eval --model hf \ - skills/skills/ai-research/evaluation-lm-evaluation-harness/SKILL.md, line 83
lm_eval --model hf \ - skills/skills/ai-research/evaluation-lm-evaluation-harness/SKILL.md, line 91
lm_eval --model hf \ - skills/skills/ai-research/evaluation-lm-evaluation-harness/SKILL.md, line 101
lm_eval --model hf \ - skills/skills/ai-research/evaluation-lm-evaluation-harness/SKILL.md, line 166
CHECKPOINT_DIR=$1 - skills/skills/ai-research/evaluation-lm-evaluation-harness/SKILL.md, line 198
os.system(f"./eval_checkpoint.sh checkpoints step-{step}") - skills/skills/ai-research/evaluation-lm-evaluation-harness/SKILL.md, line 215
os.system(f"lm_eval --model hf --model_args pretrained={checkpoint_path} ...") - skills/skills/ai-research/evaluation-lm-evaluation-harness/SKILL.md, line 258
meta-llama/Llama-2-7b-hf - skills/skills/ai-research/evaluation-lm-evaluation-harness/SKILL.md, line 270
TASKS="mmlu,gsm8k,hellaswag,truthfulqa" - skills/skills/ai-research/evaluation-lm-evaluation-harness/SKILL.md, line 345
pip install vllm - skills/skills/ai-research/evaluation-lm-evaluation-harness/SKILL.md, line 351
lm_eval --model vllm \ - skills/skills/ai-research/evaluation-lm-evaluation-harness/SKILL.md, line 363
lm_eval --model hf \ - skills/skills/ai-research/evaluation-lm-evaluation-harness/SKILL.md, line 396
lm_eval --model vllm \ - skills/skills/ai-research/evaluation-lm-evaluation-harness/SKILL.md, line 402
--num_fewshot 0 # Em vez de 5 - skills/skills/ai-research/evaluation-lm-evaluation-harness/SKILL.md, line 407
--tasks mmlu_stem # Apenas disciplinas STEM - skills/skills/ai-research/evaluation-lm-evaluation-harness/SKILL.md, line 414
--batch_size 1 # Ou --batch_size auto
and 7 more places
Can call paid AI services billed to you
The author does not mention this.
Found in 2 places.
Connects to the internet
The author does not mention this.
Found in 1 place.
Safer alternatives
Low-risk skills in the same category (other) that look related. Check that one does what you need.
Install
Paste this into your terminal. It downloads exactly the version we scanned, not whatever is newest.
if [ -e ~/.agents/skills/'evaluating-llms-harness' ] || [ -L ~/.agents/skills/'evaluating-llms-harness' ]; then echo 'A skill folder named evaluating-llms-harness already exists — not replacing it' >&2; false; else d=$(mktemp -d) && git clone --quiet --filter=blob:none --no-checkout 'https://github.com/artubss/SKILLS-CLAUDE-CODE' "$d" && git -C "$d" checkout --quiet 'ac67ed85889b15fa9e784e2a93a83f8916870698' -- 'skills/skills/ai-research/evaluation-lm-evaluation-harness' && mkdir -p ~/.agents/skills && cp -R "$d"/'skills/skills/ai-research/evaluation-lm-evaluation-harness' ~/.agents/skills/'evaluating-llms-harness' && rm -rf "$d"; fi
Codex reads ~/.agents/skills directly. For Claude Code, Cursor, Copilot, Gemini CLI, Antigravity or OpenCode, also link it. For Claude Code: ln -s ~/.agents/skills/'evaluating-llms-harness' ~/.claude/skills/'evaluating-llms-harness' (other agents: ~/.cursor/skills, ~/.copilot/skills, ~/.gemini/skills, ~/.gemini/config/skills, ~/.config/opencode/skills). Or install through the finder skill, which does both.
Badge for your README
The badge for artubss/SKILLS-CLAUDE-CODE shows the score of its lowest-scoring skill.

[](https://zalavijaysinh.github.io/skilltrust/repos/artubss/SKILLS-CLAUDE-CODE/)
Score history
| Date | Version | Score |
|---|
| 2026-11-08 | ac67ed8 | 65 |