Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/khalilbenaz/claude-skills-collectionnpx agentmods add skills/khalilbenaz/claude-skills-collection/model-fine-tunerWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/khalilbenaz/claude-skills-collection/model-fine-tuner)<a href="https://agentmods.dev/skills/khalilbenaz/claude-skills-collection/model-fine-tuner"><img src="https://agentmods.dev/badge/skills/khalilbenaz/claude-skills-collection/model-fine-tuner/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/khalilbenaz/claude-skills-collection/model-fine-tuner"><img src="https://agentmods.dev/badge/skills/khalilbenaz/claude-skills-collection/model-fine-tuner.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00116 | $0.02373 |
| Opus 5 | $0.00058 | $0.01187 |
| Sonnet 5 | $0.00023 | $0.00475 |
| Haiku 4.5 | $0.00012 | $0.00237 |
Grade A, and why
model-fine-tuner scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 246 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Model Fine-Tuner
Critères de décision : quelle méthode choisir ?
| Situation | Méthode recommandée |
|---|---|
| GPU < 24 Go, LLM > 7B | QLoRA 4-bit |
| GPU 24–80 Go, besoin de vitesse | LoRA bfloat16 |
| Petite tâche de classification | Full fine-tune BERT/DeBERTa |
| Tâche très spécifique, budget compute | IA3 ou Prefix Tuning |
| Modèle > 70B, multi-GPU | FSDP + LoRA |
| Données < 500 exemples | Prompt engineering d'abord, fine-tune ensuite |
Workflow
1. Baseline — mesurer avant de toucher quoi que ce soit
# Exemple : évaluer Mistral-7B zero-shot sur votre tâche
python eval.py --model mistralai/Mistral-7B-v0.3 --dataset data/test.jsonl --metric f1
Sans baseline chiffrée, l'amélioration est invérifiable. Loguer la baseline dans MLflow/W&B immédiatement.
2. Préparer le dataset
Format instruction/response (standard Alpaca / ChatML) :
# Format Alpaca
{"instruction": "Classe cette phrase.", "input": "Le produit est excellent.", "output": "positif"}
# Format ChatML (préféré pour les LLM modernes)
{"messages": [
{"role": "system", "content": "Tu es un classificateur de sentiment."},
{"role": "user", "content": "Le produit est excellent."},
{"role": "assistant", "content": "positif"}
]}
Validation obligatoire avant entraînement :
from datasets import load_dataset
ds = load_dataset("json", data_files={"train": "train.jsonl", "test": "test.jsonl"})
# Vérifier la distribution des labels
from collections import Counter
Counter(ds["train"]["label"])
# Vérifier les longueurs de tokens
from transformers import AutoTokenizer
tok = AutoTokenizer.from_pretrained("mistralai/Mistral-7B-v0.3")
lengths = [len(tok(x["instruction"] + x["output"])["input_ids"]) for x in ds["train"]]
print(f"max={max(lengths)}, p95={sorted(lengths)[int(0.95*len(lengths))]}")
Règle : si p95 > max_seq_length, tronquer ou filtrer — ne jamais silencieusement perdre des tokens en production.
3. Configurer LoRA / QLoRA
from transformers import AutoModelForCausalLM, BitsAndBytesConfig
from peft import LoraConfig, get_peft_model, TaskType
import torch
# QLoRA 4-bit (≈ 5 Go pour 7B)
bnb_config = BitsAndBytesConfig(
load_in_4bit=True,
bnb_4bit_quant_type="nf4",
bnb_4bit_compute_dtype=torch.bfloat16,
bnb_4bit_use_double_quant=True,
)
model = AutoModelForCausalLM.from_pretrained(
"mistralai/Mistral-7B-v0.3",
quantization_config=bnb_config,
device_map="auto",
)
lora_config = LoraConfig(
r=16, # rang : 8 (léger) → 64 (expressif)
lora_alpha=32, # scaling = alpha/r, garder alpha = 2*r
target_modules=["q_proj", "k_proj", "v_proj", "o_proj"],
lora_dropout=0.05,
bias="none",
task_type=TaskType.CAUSAL_LM,
)
model = get_peft_model(model)
model.print_trainable_parameters()
# → "trainable params: 4,194,304 || all params: 3,756,462,080 || trainable%: 0.1117"
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 246 lines · 116 tokens per session scan A 35007465e4a3
model-fine-tuner is a skill published in the GitHub repository khalilbenaz/claude-skills-collection (22 stars, last pushed 18d ago), licensed MIT. It adds 116 tokens to every session and 2,373 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
prompt-library
Curated collection of high-quality prompts for various use cases. Includes role-based prompts, task-specific templates, and prompt refinement techniques. Use when user needs prompt templates, role-play prompts, or ready-to-use prompt examples for coding, writing, analysis, or creative tasks.
edgartools
Python library for accessing, analyzing, and extracting data from SEC EDGAR filings. Use when working with SEC filings, financial statements (income statement, balance sheet, cash flow), XBRL financial data, insider trading (Form 4), institutional holdings (13F), company financials, annual/quarterly reports (10-K…
rag-patterns
RAG: embeddings, chunking, hybrid search (BM25+vector), reranking, CRAG, multi-hop. Triggers: RAG, embedding, pgvector, Qdrant, Pinecone, Weaviate, reranker, semantic search.
evaluate
Evaluates RAG retrieval and LLM-as-judge metrics (faithfulness, relevancy, context precision). Triggers: measure RAG quality, knowledge gap, RAG eval, golden dataset.
instinct-review
Reviews/promotes/removes instincts from .claude/instincts/.md. Triggers: instinct review, curate instincts, manage instincts, promote instinct.
repeat
Runs prompt/slash command on recurring interval until done or limit. Triggers: repeat, recurring task, poll status, run every N minutes, interval.