© 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
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B-IT
    7. Residual Stream - 131k
    9. 20-GEMMASCOPE-RES-131K
    11. 46877
    Prev
    Next
    INDEX
    Explanations

    occurrences of the term "claim" in various contexts, indicating a focus on legal claims and related concepts

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

    claim

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    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
    oporosis
    -0.60
     Woodford
    -0.57
     Ebers
    -0.57
     Nort
    -0.56
     Woodbury
    -0.56
    rotter
    -0.56
     superintendent
    -0.55
     Rott
    -0.53
     Gardiner
    -0.53
     forester
    -0.53
    POSITIVE LOGITS
     Claim
    1.16
    Claim
    1.13
     claim
    1.11
     CLAIM
    1.09
    claim
    1.06
    Claims
    1.03
     claims
    0.97
    CLAIM
    0.96
     CLAIMS
    0.95
     Claims
    0.95
    Activations Density 0.037%

    No Known Activations