© 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. 20-GEMMASCOPE-RES-131K
    6. 79750
    Prev
    Next
    INDEX
    Explanations

    medical or technical terms related to processes, installations, and data specifications

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

    combination, together, separately

    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
    Jîn
    -0.42
    ChrTalk
    -0.38
    PerformLayout
    -0.37
    CardContent
    -0.35
     Stake
    -0.35
     stake
    -0.32
     sho
    -0.32
     COLLINS
    -0.32
    ằm
    -0.31
     Waray
    -0.31
    POSITIVE LOGITS
    feinander
    0.65
     simultaneous
    0.61
     combination
    0.60
    combination
    0.60
     together
    0.60
     separately
    0.59
     combined
    0.59
     combinations
    0.59
     simultaneously
    0.58
     Together
    0.58
    Activations Density 0.356%

    No Known Activations