© 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. 63525
    Prev
    Next
    INDEX
    Explanations

    references to models and their specifications or interactions

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

    Model and code constructs

    np_acts-logits-general · gemini-2.5-flash-lite

    tokens that mark conversation structure and speaker turns (e.g., the "model" token and start/end-of-turn markers).

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    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
    antMatchers
    -0.45
     peng
    -0.34
     en
    -0.34
     se
    -0.33
     λε
    -0.33
     prey
    -0.31
    UrlResolution
    -0.30
    pyplot
    -0.30
     Mag
    -0.30
     Se
    -0.30
    POSITIVE LOGITS
    enderror
    0.65
    NameInMap
    0.65
     referrerpolicy
    0.63
    principalColumn
    0.58
     purpoſe
    0.57
     geschiedenis
    0.56
     myſelf
    0.56
    ſelf
    0.56
     plufieurs
    0.54
    NUMX
    0.54
    Activations Density 0.016%

    No Known Activations