© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 22-GEMMASCOPE-TRANSCODER-16K
    7. 13543
    Prev
    Next
    INDEX
    Explanations

    parts of `diff` output chunks

    oai_token-act-pair · gemini-2.0-flash

    code diffs and docker

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_22/width_16k/average_l0_15
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.ln2.hook_normalized
    Architecture
    jumprelu_transcoder
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     يتيمه
    -0.60
    исленность
    -0.57
    ::__
    -0.56
    뀜
    -0.56
     jer
    -0.56
    pertise
    -0.52
    roughly
    -0.52
    뀐
    -0.50
    Rough
    -0.50
    mės
    -0.49
    POSITIVE LOGITS
    f
    0.70
    d
    0.69
    c
    0.58
    b
    0.57
     depositphotos
    0.56
     }}"></
    0.53
     vito
    0.52
    makeText
    0.52
    lavia
    0.52
     enf
    0.51
    Activations Density 0.959%

    No Known Activations