Aquin LogoAquinLabs
Login

Supported models

Load one model at a time with aquin load model. Catalog slugs (gpt2-small, gte-small) include curated public SAEs where available. Family support lets you pass a HuggingFace repo id for other sizes in the same architecture: GPT-2, Pythia, Llama, Qwen, and common embedding encoders. Run aquin list model for the catalog and family examples.

catalog model
family model (HuggingFace repo id)

Catalog models

Curated slugs with public SAE checkpoints where available.

LLMs

Inspection, simulation, evals, and checkpoint analysis.

  • GPT-2 SmallLLMgpt2-smallPublic SAE
  • Pythia 70MLLMpythia-70mPublic HF SAE
  • Pythia 2.8BLLMpythia-2.8b
  • Llama 3.2 1BLLMllama-3.2-1b
  • Sarvam 30BLLMsarvam-30b

Embedding models

Layer drift, retrieval, embed simulation, and embedding evals. See Inspection (embedding).

  • all-MiniLM-L6Embeddingall-minilm-l6
  • GTE SmallEmbeddinggte-small
  • BGE Small ENEmbeddingbge-small-en

Model families

Pass any example HuggingFace id to aquin load model. Aquin introspects layer count and width, then runs the same tool suite as the catalog entry. Public SAE pull (aquin load sae) applies to catalog models; on family models use aquin sae train to build a temporary SAE first.

LLM families

  • GPT-2

    OpenAI GPT-2 sizes. Inspection and checkpoint tools; train your own SAE with aquin sae train.

    gpt2 · gpt2-medium · gpt2-large · gpt2-xl

  • Pythia

    EleutherAI Pythia checkpoints. Pass the full HuggingFace repo id to --model.

    EleutherAI/pythia-160m-deduped · EleutherAI/pythia-410m-deduped · EleutherAI/pythia-1.4b-deduped · EleutherAI/pythia-2.8b

  • Llama

    Meta Llama instruct checkpoints (gated; aquin login HF proxy). Catalog llama-3.2-1b includes a native public SAE.

    meta-llama/Llama-3.2-1B-Instruct · meta-llama/Llama-3.2-3B-Instruct · meta-llama/Llama-3.1-8B-Instruct

  • Qwen

    Alibaba Qwen dense checkpoints (Qwen2, Qwen2.5, Qwen3). Loaded via HuggingFace shim; train your own SAE with aquin sae train.

    Qwen/Qwen3-8B · Qwen/Qwen2.5-7B-Instruct · Qwen/Qwen2.5-0.5B-Instruct

Embedding families

  • BERT-style encoders

    Sentence-transformers, GTE, BGE, and E5 families. Full embed inspection suite after aquin load model.

    thenlper/gte-base · BAAI/bge-base-en-v1.5 · sentence-transformers/all-mpnet-base-v2 · intfloat/e5-small-v2

SAEs & checkpoints

  • One model in VRAM at a time. Run aquin load model again to switch.
  • Catalog: feature tools need a public SAE. Run aquin list sae, then aquin load sae <model-l{n}>.
  • Family models: train with aquin sae train, then aquin load sae --user <name> before inspect.
  • Fine-tuned weights stay on disk. Use --checkpoint while the base model id stays loaded.
  • SAE catalog: aquin list sae · SAE training · Checkpoint tools