© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-2B
    7. Residual Stream - 16k
    9. 12-GEMMASCOPE-RES-16K
    11. 650
    Prev
    Next
    INDEX
    Explanations

    formal questions and answers related to opinions and statements

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

    positions of speaker labels and colons that mark turns in a transcript (e.g., "Q:", "A:", or other colon-delimited speaker markers).

    oai_token-act-pair · gpt-5-miniTriggered by @phamvanluong2003

    markers of interview or transcript-style Q&A dialogue, especially speaker labels with colons and quoted question/answer turns.

    oai_token-act-pair · gpt-5Triggered by @areeb0102

    dialogue markers in interview or conversational exchanges.

    oai_token-act-pair · deepseek-r1Triggered by @areeb0102

    question-and-answer or interview dialogue formatting and structure.

    oai_token-act-pair · claude-4-5-haikuTriggered by @areeb0102
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 12-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_12/width_16k/average_l0_82
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.12.hook_resid_post
    Hook Layer
    12
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     يتيمه
    -0.96
    DockStyle
    -0.86
    StructEnd
    -0.82
    ientôt
    -0.79
    +#+
    -0.79
    enapa
    -0.76
     TAMBÉM
    -0.76
     NavController
    -0.76
    oodoo
    -0.76
    tvguidetime
    -0.74
    POSITIVE LOGITS
    ↵↵
    0.64
    I
    0.59
     I
    0.55
    Yes
    0.47
     Certainly
    0.43
     certainly
    0.43
    Certainly
    0.43
     tecnici
    0.43
     thèse
    0.42
    Well
    0.42
    Activations Density 0.085%

    No Known Activations