© 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. 86062
    Prev
    Next
    INDEX
    Explanations

    terms related to enthusiasts and aficionados of specific interests or subjects

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

    enthusiasts and lovers

    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
    astify
    -0.42
     AssemblyCulture
    -0.40
     Ticket
    -0.40
     AssemblyCompany
    -0.40
    matica
    -0.37
     receipt
    -0.36
    __':
    -0.34
     ola
    -0.34
     Duet
    -0.34
    Ticket
    -0.33
    POSITIVE LOGITS
     vegetarians
    0.53
     enthusiasts
    0.52
     nerds
    0.50
     âmes
    0.50
     lovers
    0.49
     devotees
    0.48
     souls
    0.48
     Roskov
    0.47
     enthusiast
    0.47
    igshid
    0.47
    Activations Density 0.020%

    No Known Activations