tokens&
For enterprises
Submit
Sign in
tokens&

Build better AI stacks, claim useful opportunities, and give AI infrastructure companies a source-labeled adoption readout they can trust.

For buildersFor enterprises

Product

  • For builders
  • Category rankings
  • Startup credits and perks
  • Agent Skills
  • Platform
  • Submit project, tool, product, or perk

Enterprise

  • Start free company workspace

Community

  • Community
  • Newsletter
  • Events
Xin

© 2026 tokensand, LLC. All rights reserved.

  • Terms
  • Privacy
  • Security
  • Data Processing
  • Status
Agent Skills/Sentence Transformers trainer
Hugging FaceModelsSKILL.mdVerified source

Agent Skill

Sentence Transformers trainer

Fine-tune embedding models for retrieval, similarity, clustering, and classification.

Install this skillView repository

Vendor-authored source · Apache-2.0 license.

Raw SKILL.mdInstall the Tokens& Agent Pack

Skill specification

Declared by Hugging Face in the package front matter. Trigger conditions are what the coding agent matches on before it loads the skill.

View package fields
Sentence Transformers trainer SKILL.md front matter fields
Skill nametrain-sentence-transformers
Trigger conditionsTrain or fine-tune sentence-transformers models across `SentenceTransformer` (bi-encoder, dense or static embedding model for retrieval, similarity, clustering, classification, paraphrase mining, dedup, multimodal), `CrossEncoder` (reranker, pair scoring for two-stage retrieval / pair classification), `SparseEncoder` (SPLADE, sparse embedding model for learned-sparse retrieval), and `MultiVectorEncoder` (ColBERT / late-interaction, per-token embeddings scored with MaxSim). Covers loss selection, hard-negative mining, evaluators, distillation, LoRA, Matryoshka, and Hugging Face Hub publishing. Use for any sentence-transformers training task.

Install train-sentence-transformers

In a terminal with Node.js, npm and Git, run the command for your agent. The Skills CLI installs the complete package directory, including referenced files within it. Review its install prompt, then start a new agent session. A skill package does not set up an MCP server connection.

Claude Code

.claude/skills/train-sentence-transformers/SKILL.md

Project skills are committed with the repo. Use the user directory for a personal install across every project.

Project install

npx skills add 'https://github.com/huggingface/skills/tree/main/skills/train-sentence-transformers' --skill 'train-sentence-transformers' --agent 'claude-code'
Install for all projects instead

Personal install

npx skills add 'https://github.com/huggingface/skills/tree/main/skills/train-sentence-transformers' --skill 'train-sentence-transformers' --agent 'claude-code' --global

Codex

.agents/skills/train-sentence-transformers/SKILL.md

Codex reads `.agents/skills/` as its primary location, which is also the cross-platform default other clients honour.

Project install

npx skills add 'https://github.com/huggingface/skills/tree/main/skills/train-sentence-transformers' --skill 'train-sentence-transformers' --agent 'codex'
Install for all projects instead

Personal install

npx skills add 'https://github.com/huggingface/skills/tree/main/skills/train-sentence-transformers' --skill 'train-sentence-transformers' --agent 'codex' --global

Cursor

.agents/skills/train-sentence-transformers/SKILL.md

Cursor also loads `.agents/skills/`, `.claude/skills/`, and `.codex/skills/`, so one committed copy can serve several clients.

Project install

npx skills add 'https://github.com/huggingface/skills/tree/main/skills/train-sentence-transformers' --skill 'train-sentence-transformers' --agent 'cursor'
Install for all projects instead

Personal install

npx skills add 'https://github.com/huggingface/skills/tree/main/skills/train-sentence-transformers' --skill 'train-sentence-transformers' --agent 'cursor' --global

Gemini CLI

.agents/skills/train-sentence-transformers/SKILL.md

Gemini CLI reads `.agents/skills/` first when both directories exist.

Project install

npx skills add 'https://github.com/huggingface/skills/tree/main/skills/train-sentence-transformers' --skill 'train-sentence-transformers' --agent 'gemini-cli'
Install for all projects instead

Personal install

npx skills add 'https://github.com/huggingface/skills/tree/main/skills/train-sentence-transformers' --skill 'train-sentence-transformers' --agent 'gemini-cli' --global

SKILL.md

View raw source

Published by Hugging Face under Apache-2.0. Rendered from the package in github.com/huggingface/skills/tree/main/skills/train-sentence-transformers.

Read full skill instructions

Train a sentence-transformers Model

This SKILL.md is a router, not a manual. It tells you which references and example scripts to load for your task. The actual content (recommended losses, evaluators, training-script structure, model selection, training-arg knobs, troubleshooting) lives in references/ and scripts/.

Do not synthesize a training script from this file alone. Open the per-type production template (scripts/train_<type>_example.py) and copy it as your starting point. The templates contain load-bearing scaffolding (autocast helper, model-card class, logger silencing list, force=True, seed, TF32, version-compatible imports, named-evaluator metric handling) that prior agent runs have repeatedly missed when rolling their own from a synthesized snippet.

1. Identify the model type

TagClassWhat it doesWhen to pick
[SentenceTransformer]SentenceTransformer (bi-encoder)Maps each input to a fixed-dim dense vectorRetrieval, similarity, clustering, classification, paraphrase mining, dedup
[CrossEncoder]CrossEncoder (reranker)Scores (query, passage) pairs jointlyTwo-stage retrieval (rerank top-100 from bi-encoder), pair classification
[SparseEncoder]SparseEncoder (SPLADE)Sparse vectors over the vocabularyLearned-sparse retrieval, inverted-index backends (Elasticsearch / OpenSearch / Lucene)
[MultiVectorEncoder]MultiVectorEncoder (ColBERT)One embedding per token, scored with MaxSimLate-interaction retrieval, recall gains over bi-encoders at higher storage cost, multimodal (ColPali / ColQwen2)

Tiebreakers when the request is ambiguous: "embedding model" / "vector search" / "similarity" → [SentenceTransformer]. "rerank" / "ranker" / "two-stage" → [CrossEncoder]. "SPLADE" / "sparse" / "inverted index" → [SparseEncoder]. "ColBERT" / "late interaction" / "multi-vector" / "MaxSim" / "ColPali" / "ColQwen" → [MultiVectorEncoder]. If still unclear, ask.

2. Required reading

Read these in full before writing any code. Do not triage by perceived relevance.

Per-type: always required

[SentenceTransformer]

  • references/losses_sentence_transformer.md: loss-to-data-shape mapping, BatchSamplers.NO_DUPLICATES requirement for MNRL-family, Cached* ↔ gradient_checkpointing incompatibility.
  • references/evaluators_sentence_transformer.md: evaluator-to-task mapping, metric_for_best_model key construction (named vs unnamed), per-evaluator primary_metric values.
  • references/model_architectures.md: encoder vs decoder vs static vs Router pipelines, pooling rules (mean / cls / lasttoken), auto-mean-pooling behavior for fresh-start MLM bases.
  • scripts/train_sentence_transformer_example.py: production template. Copy this as your starting point.

[CrossEncoder]

  • references/losses_cross_encoder.md: pointwise / pairwise / listwise / distillation, pos_weight derivation, activation_fn=Identity() mandatory for non-BCE losses (silent eval-rank collapse otherwise).
  • references/evaluators_cross_encoder.md: CrossEncoderRerankingEvaluator recipe, named-evaluator key format eval_{name}_{primary_metric}.
  • scripts/train_cross_encoder_example.py: production template. Copy this as your starting point.

[SparseEncoder]

  • references/losses_sparse_encoder.md: SpladeLoss wrapper requirement, FLOPS regularizer weights, smoke-test active-dim ramp behavior.
  • references/evaluators_sparse_encoder.md: SparseNanoBEIREvaluator (English-only) and the in-domain alternative, eval_{name}_{primary_metric} key format.
  • scripts/train_sparse_encoder_example.py: production template. Copy this as your starting point.

[MultiVectorEncoder]

  • references/losses_multi_vector_encoder.md: MaxSim scoring, scale choice per scoring mode (scale=1.0 for MaxSim, roughly the average query length for MeanMaxSim), MNRL / CachedMNRL / MarginMSE / DistillKLDiv, XTR-vs-ColBERT scoring, CachedMNRL ↔ gradient_checkpointing incompatibility.
  • references/evaluators_multi_vector_encoder.md: MultiVectorNanoBEIREvaluator (English-only) and the in-domain alternative, eval_NanoBEIR_mean_maxsim_ndcg@10 key format, distillation-eval spearman variant.
  • scripts/train_multi_vector_encoder_example.py: production template. Copy this as your starting point.

Cross-cutting: always required (regardless of task)

  • references/training_args.md: TrainingArguments knobs, precision rules (load fp32 + autocast bf16/fp16, never torch_dtype=bfloat16), warmup_steps (float) vs deprecated warmup_ratio, save_steps must be a multiple of eval_steps for load_best_model_at_end, schedulers, HPO, tracker, resume, hub-push variants.
  • references/dataset_formats.md: column-matching rules (label name auto-detection, column-order-not-name), reshaping recipes, hard-negative mining options.
  • references/base_model_selection.md: discovery commands, per-type model namespaces, ModernBERT-family max_seq_length=8192 trap, datasets >= 4 script-loader rejection, non-English starting-point shortcuts.
  • references/troubleshooting.md: symptom-indexed failure recipes. Skim the section headings on every run, even a healthy one. The "Metrics don't improve" and "Hub push fails" entries cover bugs that bite frequently and are cheaper to recognize before they fire than to debug after.

Cross-cutting: load when applicable

  • references/hardware_guide.md: VRAM sizing, multi-GPU, FSDP / DeepSpeed, HF Jobs flavors. Required for >24GB models, multi-GPU, or HF Jobs runs.
  • references/hf_jobs_execution.md: required when running on HF Jobs.
  • references/prompts_and_instructions.md: required when using prompt-tuned bases (E5, BGE, GTE, Qwen3-Embedding, Instructor, Nomic, etc.) or adding query: / passage: style prefixes.

Variant scripts (open when the task matches)

  • [SentenceTransformer] scripts/train_sentence_transformer_<matryoshka|multi_dataset|with_lora|distillation|make_multilingual|static_embedding>_example.py.
  • [CrossEncoder] scripts/train_cross_encoder_<distillation|listwise>_example.py.
  • [SparseEncoder] scripts/train_sparse_encoder_distillation_example.py.
  • Hard-negative mining CLI: scripts/mine_hard_negatives.py.

3. Defaults

Override only if the user specifies otherwise:

  • Local execution. Pitch HF Jobs only if local hardware can't fit the job.
  • Single run. After it completes, propose experimentation if the user would benefit (weak/marginal verdict, "see how high you can push it" framing, etc.). Iteration rules in references/training_args.md (Experimentation section).
  • Public Hub push at end-of-run, wrapped in try-except. On HF Jobs (ephemeral env) ALSO enable in-trainer push (push_to_hub=True + hub_strategy="every_save"). Details in references/hf_jobs_execution.md.

4. Constraints the produced script must satisfy

These are non-negotiable contracts. Implementation lives in the production templates and references. Do not reinvent.

  • Capture the pre-training evaluator score as baseline_eval before trainer.train().
  • Emit a single end-of-run line: VERDICT: WIN|MARGINAL|REGRESSION | score=... | baseline=... | delta=.... A monitor scrapes for this.
  • Silence httpx, httpcore, huggingface_hub, urllib3, filelock, fsspec to WARNING (otherwise HF download URLs flood the agent's context).
  • Tee logs to logs/{RUN_NAME}.log.
  • End with model.push_to_hub(...) wrapped in try/except.
  • Smoke-test before any long run (max_steps=1 + tiny dataset slice). The production templates show one common pattern (SMOKE_TEST env var).
  • [CrossEncoder] Include EarlyStoppingCallback(patience>=3). CE rerankers often peak mid-training and regress.
  • [SparseEncoder] Log query_active_dims

5. Workflow

  1. Identify the model type (§1). Ask if ambiguous.
  2. Load the §2 required-reading files for that type.
  3. Open scripts/train_<type>_example.py and copy it as your starting point.
  4. Replace MODEL_NAME, DATASET_NAME, RUN_NAME, the loss, and the evaluator with the user's task. Cross-check loss/data-shape match against references/losses_<type>.md. Cross-check the metric_for_best_model key against references/evaluators_<type>.md (named evaluators format the key as eval_{name}_{primary_metric}).
  5. Smoke-test (max_steps=1).
  6. Run.
  7. After the run, append to logs/experiments.md and propose iteration if the verdict is weak/marginal.

Prerequisites

pip install "sentence-transformers[train]>=5.0"        # add [train,image] / [audio] / [video] for [SentenceTransformer] multimodal
                                                       # [MultiVectorEncoder] requires >=6.0
pip install trackio                                    # optional tracker (or wandb / tensorboard / mlflow)
hf auth login                                          # or set HF_TOKEN with write scope (for Hub push)

GPU strongly recommended. CPU works only for demos and [SentenceTransformer] StaticEmbedding.

More Hugging Face Agent Skills

All Agent Skills

Community model evals

Run inspect-ai and lighteval evaluations for model selection, regressions, and benchmarks.

Models

Dataset Viewer workflows

Fetch dataset metadata, split rows, filtered samples, and viewer API results for data tasks.

Models

Gradio app builder

Build and edit Gradio apps, demos, chatbots, layouts, and event listeners.

Frontend

Hub tool builder

Create scripts and tools that chain Hugging Face Hub API data safely.

Agents

Hugging Face CLI

Package model, dataset, and Space operations into a repeatable agent workflow.

Models

Hugging Face MCP

Use Hugging Face Hub MCP tools for models, datasets, Spaces, papers, and compute jobs.

Agents

/
corpus_active_dims
on the verdict line. High nDCG with collapsed sparsity is not a win. The keys come back name-prefixed (e.g.
..._query_active_dims
). Use suffix matching to pluck them. See the SPARSE production template for the exact pattern.
  • [MultiVectorEncoder] Match scale to the scoring mode on any MNRL-family loss: near 1.0 for unnormalized MaxSim (do not copy scale=20.0 from bi-encoder MNRL), roughly the average query length with length-normalized MeanMaxSim, since each score is divided by its query's token count. XTRScores is a train-only similarity_fct: the evaluators reject it, so evaluation always scores with MaxSim, including for XTR-trained models.