© 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. 31-GEMMASCOPE-RES-131K
    6. 69450
    Prev
    Next
    INDEX
    Explanations

    references to claims and assertions related to territories or jurisdictions

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

    claim responsibility or territory

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     SDLK
    -0.34
    SuccessListener
    -0.31
     пого
    -0.31
     fundido
    -0.31
    exitRule
    -0.30
     reda
    -0.30
     espoir
    -0.30
     ſer
    -0.29
     onPostExecute
    -0.29
    ImageContext
    -0.28
    POSITIVE LOGITS
     claim
    1.39
     territorial
    1.34
     territory
    1.34
     Claim
    1.29
     claiming
    1.28
     claimed
    1.28
     CLAIM
    1.27
    claim
    1.23
    Claim
    1.20
    territory
    1.20
    Activations Density 0.129%

    No Known Activations