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. 61899
    Prev
    Next
    INDEX
    Explanations

    references to unlimited access or bandwidth

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

    unlimited proxy for pc

    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
     Algunos
    -0.40
    CardHeader
    -0.40
    WriteLiteral
    -0.40
    section
    -0.37
    case
    -0.37
     a
    -0.36
    key
    -0.35
    jenigen
    -0.35
    _
    -0.35
    geordneten
    -0.35
    POSITIVE LOGITS
     Unlimited
    1.16
    Unlimited
    1.13
     unlimited
    1.10
    unlimited
    1.04
     illimit
    0.92
     ilimit
    0.86
     limitless
    0.80
     unbounded
    0.79
    无限
    0.73
    無限
    0.72
    Activations Density 0.003%

    No Known Activations