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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 85250
    Prev
    Next
    INDEX
    Explanations

    key terms related to jobs, future prospects, and actions associated with employment

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

    "th century" / "New York" / "Carnegie Mellon"

    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
     Détails
    -0.40
     fraî
    -0.38
    cteurs
    -0.38
    於是
    -0.38
     frescas
    -0.36
    Fresh
    -0.36
     télécharge
    -0.35
    horabuena
    -0.35
     délai
    -0.35
    FRESH
    -0.35
    POSITIVE LOGITS
     surla
    0.68
    featureID
    0.64
    ViewFeatures
    0.63
    AndEndTag
    0.59
    EndContext
    0.58
    matchCondition
    0.56
    RTSC
    0.54
    IUrlHelper
    0.53
    principalColumn
    0.53
    errHandler
    0.53
    Activations Density 0.044%

    No Known Activations