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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 20-GEMMASCOPE-RES-16K
    6. 2136
    Prev
    Next
    INDEX
    Explanations

    references to scientific concepts and inquiries

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    references to specific quantitative aspects or metrics in technical or financial writing.

    oai_token-act-pair · claude-3-5-haiku-20241022Triggered by @neilrathi

    plural nouns and objects that follow verbs and prepositions.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    plural common nouns, especially abstract or content-heavy items within noun phrases.

    oai_token-act-pair · gpt-5Triggered by @jyhe0408

    nouns referring to abstract concepts, technical subjects, or specific entities in formal or academic contexts.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.20.hook_resid_post
    Hook Layer
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    brenner
    -0.51
    뀜
    -0.47
    nolia
    -0.43
    ſelben
    -0.43
     annulus
    -0.43
    Mol
    -0.43
    assembly
    -0.42
     CascadeType
    -0.42
     dezelve
    -0.42
     administering
    -0.41
    POSITIVE LOGITS
     such
    0.52
     similar
    0.51
     solchen
    0.49
     solche
    0.48
     GoogleFonts
    0.47
     liknande
    0.47
     muun
    0.44
    providedIn
    0.44
     صوتيه
    0.44
    อื่น
    0.43
    Activations Density 0.765%

    No Known Activations