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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 9-GEMMASCOPE-RES-131K
    6. 68526
    Prev
    Next
    INDEX
    Explanations

    instances of the word "bent" and its variations, indicating a focus on curvature or inclination

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

    Bent's Fort

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.9.hook_resid_post
    Hook Layer
    9
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Händ
    -0.55
     مشين
    -0.54
     Shaffer
    -0.53
     للاسماء
    -0.50
     Penga
    -0.50
     Barrios
    -0.49
    spesies
    -0.48
    天王
    -0.48
    ecma
    -0.47
    providedIn
    -0.47
    POSITIVE LOGITS
     bent
    2.11
     Bent
    2.05
    Bent
    1.96
    bent
    1.96
     Benton
    1.14
     bend
    1.05
     Bend
    1.02
    bend
    1.00
     bends
    0.98
     BEND
    0.92
    Activations Density 0.007%

    No Known Activations