Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 75017
    Prev
    Next
    INDEX
    Explanations

    references to fan engagement and testimonials related to media or entertainment

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

    interviewees and professionals discussing their work

    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
    CloseOperation
    -0.72
     HttpNotFound
    -0.66
    featureID
    -0.63
    InputTagHelper
    -0.55
     DriverManager
    -0.54
    LabelTagHelper
    -0.54
    aarrggbb
    -0.52
    pexpr
    -0.51
    yntaxException
    -0.51
     intStringLen
    -0.50
    POSITIVE LOGITS
     interviewed
    0.60
     interview
    0.60
     interviews
    0.57
     interviewees
    0.53
     opinião
    0.49
    Entrevista
    0.47
     voces
    0.46
     INTERVIEW
    0.46
    Interview
    0.45
     opiniones
    0.44
    Activations Density 0.045%

    No Known Activations