MCP HubMCP Hub
스킬 목록으로 돌아가기

evaluating-llms-harness

zechenzhangAGI
업데이트됨 Today
1362 조회
62
2
62
GitHub에서 보기
테스팅aitestingapi

정보

이 Claude Skill은 MMLU, GSM8K를 포함한 60개 이상의 표준화된 학술 과제에서 LLM 성능을 벤치마크하기 위해 lm-evaluation-harness를 실행합니다. 개발자들이 모델 품질을 비교하고, 학습 진행 상황을 추적하거나 학술 결과를 보고할 수 있도록 설계되었습니다. 이 도구는 HuggingFace와 vLLM 모델을 포함한 다양한 백엔드를 지원합니다.

빠른 설치

Claude Code

추천
기본
npx skills add zechenzhangAGI/AI-research-SKILLs
플러그인 명령대체
/plugin add https://github.com/zechenzhangAGI/AI-research-SKILLs
Git 클론대체
git clone https://github.com/zechenzhangAGI/AI-research-SKILLs.git ~/.claude/skills/evaluating-llms-harness

Claude Code에서 이 명령을 복사하여 붙여넣어 스킬을 설치하세요

문서

lm-evaluation-harness - LLM Benchmarking

Quick start

lm-evaluation-harness evaluates LLMs across 60+ academic benchmarks using standardized prompts and metrics.

Installation:

pip install lm-eval

Evaluate any HuggingFace model:

lm_eval --model hf \
  --model_args pretrained=meta-llama/Llama-2-7b-hf \
  --tasks mmlu,gsm8k,hellaswag \
  --device cuda:0 \
  --batch_size 8

View available tasks:

lm_eval --tasks list

Common workflows

Workflow 1: Standard benchmark evaluation

Evaluate model on core benchmarks (MMLU, GSM8K, HumanEval).

Copy this checklist:

Benchmark Evaluation:
- [ ] Step 1: Choose benchmark suite
- [ ] Step 2: Configure model
- [ ] Step 3: Run evaluation
- [ ] Step 4: Analyze results

Step 1: Choose benchmark suite

Core reasoning benchmarks:

  • MMLU (Massive Multitask Language Understanding) - 57 subjects, multiple choice
  • GSM8K - Grade school math word problems
  • HellaSwag - Common sense reasoning
  • TruthfulQA - Truthfulness and factuality
  • ARC (AI2 Reasoning Challenge) - Science questions

Code benchmarks:

  • HumanEval - Python code generation (164 problems)
  • MBPP (Mostly Basic Python Problems) - Python coding

Standard suite (recommended for model releases):

--tasks mmlu,gsm8k,hellaswag,truthfulqa,arc_challenge

Step 2: Configure model

HuggingFace model:

lm_eval --model hf \
  --model_args pretrained=meta-llama/Llama-2-7b-hf,dtype=bfloat16 \
  --tasks mmlu \
  --device cuda:0 \
  --batch_size auto  # Auto-detect optimal batch size

Quantized model (4-bit/8-bit):

lm_eval --model hf \
  --model_args pretrained=meta-llama/Llama-2-7b-hf,load_in_4bit=True \
  --tasks mmlu \
  --device cuda:0

Custom checkpoint:

lm_eval --model hf \
  --model_args pretrained=/path/to/my-model,tokenizer=/path/to/tokenizer \
  --tasks mmlu \
  --device cuda:0

Step 3: Run evaluation

# Full MMLU evaluation (57 subjects)
lm_eval --model hf \
  --model_args pretrained=meta-llama/Llama-2-7b-hf \
  --tasks mmlu \
  --num_fewshot 5 \  # 5-shot evaluation (standard)
  --batch_size 8 \
  --output_path results/ \
  --log_samples  # Save individual predictions

# Multiple benchmarks at once
lm_eval --model hf \
  --model_args pretrained=meta-llama/Llama-2-7b-hf \
  --tasks mmlu,gsm8k,hellaswag,truthfulqa,arc_challenge \
  --num_fewshot 5 \
  --batch_size 8 \
  --output_path results/llama2-7b-eval.json

Step 4: Analyze results

Results saved to results/llama2-7b-eval.json:

{
  "results": {
    "mmlu": {
      "acc": 0.459,
      "acc_stderr": 0.004
    },
    "gsm8k": {
      "exact_match": 0.142,
      "exact_match_stderr": 0.006
    },
    "hellaswag": {
      "acc_norm": 0.765,
      "acc_norm_stderr": 0.004
    }
  },
  "config": {
    "model": "hf",
    "model_args": "pretrained=meta-llama/Llama-2-7b-hf",
    "num_fewshot": 5
  }
}

Workflow 2: Track training progress

Evaluate checkpoints during training.

Training Progress Tracking:
- [ ] Step 1: Set up periodic evaluation
- [ ] Step 2: Choose quick benchmarks
- [ ] Step 3: Automate evaluation
- [ ] Step 4: Plot learning curves

Step 1: Set up periodic evaluation

Evaluate every N training steps:

#!/bin/bash
# eval_checkpoint.sh

CHECKPOINT_DIR=$1
STEP=$2

lm_eval --model hf \
  --model_args pretrained=$CHECKPOINT_DIR/checkpoint-$STEP \
  --tasks gsm8k,hellaswag \
  --num_fewshot 0 \  # 0-shot for speed
  --batch_size 16 \
  --output_path results/step-$STEP.json

Step 2: Choose quick benchmarks

Fast benchmarks for frequent evaluation:

  • HellaSwag: ~10 minutes on 1 GPU
  • GSM8K: ~5 minutes
  • PIQA: ~2 minutes

Avoid for frequent eval (too slow):

  • MMLU: ~2 hours (57 subjects)
  • HumanEval: Requires code execution

Step 3: Automate evaluation

Integrate with training script:

# In training loop
if step % eval_interval == 0:
    model.save_pretrained(f"checkpoints/step-{step}")

    # Run evaluation
    os.system(f"./eval_checkpoint.sh checkpoints step-{step}")

Or use PyTorch Lightning callbacks:

from pytorch_lightning import Callback

class EvalHarnessCallback(Callback):
    def on_validation_epoch_end(self, trainer, pl_module):
        step = trainer.global_step
        checkpoint_path = f"checkpoints/step-{step}"

        # Save checkpoint
        trainer.save_checkpoint(checkpoint_path)

        # Run lm-eval
        os.system(f"lm_eval --model hf --model_args pretrained={checkpoint_path} ...")

Step 4: Plot learning curves

import json
import matplotlib.pyplot as plt

# Load all results
steps = []
mmlu_scores = []

for file in sorted(glob.glob("results/step-*.json")):
    with open(file) as f:
        data = json.load(f)
        step = int(file.split("-")[1].split(".")[0])
        steps.append(step)
        mmlu_scores.append(data["results"]["mmlu"]["acc"])

# Plot
plt.plot(steps, mmlu_scores)
plt.xlabel("Training Step")
plt.ylabel("MMLU Accuracy")
plt.title("Training Progress")
plt.savefig("training_curve.png")

Workflow 3: Compare multiple models

Benchmark suite for model comparison.

Model Comparison:
- [ ] Step 1: Define model list
- [ ] Step 2: Run evaluations
- [ ] Step 3: Generate comparison table

Step 1: Define model list

# models.txt
meta-llama/Llama-2-7b-hf
meta-llama/Llama-2-13b-hf
mistralai/Mistral-7B-v0.1
microsoft/phi-2

Step 2: Run evaluations

#!/bin/bash
# eval_all_models.sh

TASKS="mmlu,gsm8k,hellaswag,truthfulqa"

while read model; do
    echo "Evaluating $model"

    # Extract model name for output file
    model_name=$(echo $model | sed 's/\//-/g')

    lm_eval --model hf \
      --model_args pretrained=$model,dtype=bfloat16 \
      --tasks $TASKS \
      --num_fewshot 5 \
      --batch_size auto \
      --output_path results/$model_name.json

done < models.txt

Step 3: Generate comparison table

import json
import pandas as pd

models = [
    "meta-llama-Llama-2-7b-hf",
    "meta-llama-Llama-2-13b-hf",
    "mistralai-Mistral-7B-v0.1",
    "microsoft-phi-2"
]

tasks = ["mmlu", "gsm8k", "hellaswag", "truthfulqa"]

results = []
for model in models:
    with open(f"results/{model}.json") as f:
        data = json.load(f)
        row = {"Model": model.replace("-", "/")}
        for task in tasks:
            # Get primary metric for each task
            metrics = data["results"][task]
            if "acc" in metrics:
                row[task.upper()] = f"{metrics['acc']:.3f}"
            elif "exact_match" in metrics:
                row[task.upper()] = f"{metrics['exact_match']:.3f}"
        results.append(row)

df = pd.DataFrame(results)
print(df.to_markdown(index=False))

Output:

| Model                  | MMLU  | GSM8K | HELLASWAG | TRUTHFULQA |
|------------------------|-------|-------|-----------|------------|
| meta-llama/Llama-2-7b  | 0.459 | 0.142 | 0.765     | 0.391      |
| meta-llama/Llama-2-13b | 0.549 | 0.287 | 0.801     | 0.430      |
| mistralai/Mistral-7B   | 0.626 | 0.395 | 0.812     | 0.428      |
| microsoft/phi-2        | 0.560 | 0.613 | 0.682     | 0.447      |

Workflow 4: Evaluate with vLLM (faster inference)

Use vLLM backend for 5-10x faster evaluation.

vLLM Evaluation:
- [ ] Step 1: Install vLLM
- [ ] Step 2: Configure vLLM backend
- [ ] Step 3: Run evaluation

Step 1: Install vLLM

pip install vllm

Step 2: Configure vLLM backend

lm_eval --model vllm \
  --model_args pretrained=meta-llama/Llama-2-7b-hf,tensor_parallel_size=1,dtype=auto,gpu_memory_utilization=0.8 \
  --tasks mmlu \
  --batch_size auto

Step 3: Run evaluation

vLLM is 5-10× faster than standard HuggingFace:

# Standard HF: ~2 hours for MMLU on 7B model
lm_eval --model hf \
  --model_args pretrained=meta-llama/Llama-2-7b-hf \
  --tasks mmlu \
  --batch_size 8

# vLLM: ~15-20 minutes for MMLU on 7B model
lm_eval --model vllm \
  --model_args pretrained=meta-llama/Llama-2-7b-hf,tensor_parallel_size=2 \
  --tasks mmlu \
  --batch_size auto

When to use vs alternatives

Use lm-evaluation-harness when:

  • Benchmarking models for academic papers
  • Comparing model quality across standard tasks
  • Tracking training progress
  • Reporting standardized metrics (everyone uses same prompts)
  • Need reproducible evaluation

Use alternatives instead:

  • HELM (Stanford): Broader evaluation (fairness, efficiency, calibration)
  • AlpacaEval: Instruction-following evaluation with LLM judges
  • MT-Bench: Conversational multi-turn evaluation
  • Custom scripts: Domain-specific evaluation

Common issues

Issue: Evaluation too slow

Use vLLM backend:

lm_eval --model vllm \
  --model_args pretrained=model-name,tensor_parallel_size=2

Or reduce fewshot examples:

--num_fewshot 0  # Instead of 5

Or evaluate subset of MMLU:

--tasks mmlu_stem  # Only STEM subjects

Issue: Out of memory

Reduce batch size:

--batch_size 1  # Or --batch_size auto

Use quantization:

--model_args pretrained=model-name,load_in_8bit=True

Enable CPU offloading:

--model_args pretrained=model-name,device_map=auto,offload_folder=offload

Issue: Different results than reported

Check fewshot count:

--num_fewshot 5  # Most papers use 5-shot

Check exact task name:

--tasks mmlu  # Not mmlu_direct or mmlu_fewshot

Verify model and tokenizer match:

--model_args pretrained=model-name,tokenizer=same-model-name

Issue: HumanEval not executing code

Install execution dependencies:

pip install human-eval

Enable code execution:

lm_eval --model hf \
  --model_args pretrained=model-name \
  --tasks humaneval \
  --allow_code_execution  # Required for HumanEval

Advanced topics

Benchmark descriptions: See references/benchmark-guide.md for detailed description of all 60+ tasks, what they measure, and interpretation.

Custom tasks: See references/custom-tasks.md for creating domain-specific evaluation tasks.

API evaluation: See references/api-evaluation.md for evaluating OpenAI, Anthropic, and other API models.

Multi-GPU strategies: See references/distributed-eval.md for data parallel and tensor parallel evaluation.

Hardware requirements

  • GPU: NVIDIA (CUDA 11.8+), works on CPU (very slow)
  • VRAM:
    • 7B model: 16GB (bf16) or 8GB (8-bit)
    • 13B model: 28GB (bf16) or 14GB (8-bit)
    • 70B model: Requires multi-GPU or quantization
  • Time (7B model, single A100):
    • HellaSwag: 10 minutes
    • GSM8K: 5 minutes
    • MMLU (full): 2 hours
    • HumanEval: 20 minutes

Resources

GitHub 저장소

zechenzhangAGI/AI-research-SKILLs
경로: 11-evaluation/lm-evaluation-harness
aiai-researchclaudeclaude-codeclaude-skillscodex

연관 스킬

content-collections

메타

This skill provides a production-tested setup for Content Collections, a TypeScript-first tool that transforms Markdown/MDX files into type-safe data collections with Zod validation. Use it when building blogs, documentation sites, or content-heavy Vite + React applications to ensure type safety and automatic content validation. It covers everything from Vite plugin configuration and MDX compilation to deployment optimization and schema validation.

스킬 보기

polymarket

메타

This skill enables developers to build applications with the Polymarket prediction markets platform, including API integration for trading and market data. It also provides real-time data streaming via WebSocket to monitor live trades and market activity. Use it for implementing trading strategies or creating tools that process live market updates.

스킬 보기

creating-opencode-plugins

메타

This skill helps developers create OpenCode plugins that hook into 25+ event types like commands, files, and LSP operations. It provides the plugin structure, event API specifications, and implementation patterns for JavaScript/TypeScript modules. Use it when you need to intercept, monitor, or extend the OpenCode AI assistant's lifecycle with custom event-driven logic.

스킬 보기

himalaya-email-manager

커뮤니케이션

This Claude Skill enables email management through the Himalaya CLI tool using IMAP. It allows developers to search, summarize, and delete emails from an IMAP account with natural language queries. Use it for automated email workflows like getting daily summaries or performing batch operations directly from Claude.

스킬 보기