© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 9-GEMMASCOPE-RES-131K
    6. 11995
    Prev
    Next
    INDEX
    Explanations

    occurrences of certain high-value tokens, indicating a focus on the structural or formal aspects of content

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

    code files and programming languages

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k
    Configuration
    google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.9.hook_resid_post
    Hook Layer
    9
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Personendaten
    -0.71
    はじめに
    -0.64
    adpleegd
    -0.64
    AddTagHelper
    -0.62
    CCER
    -0.60
    endphp
    -0.60
    ksikon
    -0.59
    躇
    -0.59
     opium
    -0.59
    发表于
    -0.58
    POSITIVE LOGITS
     StatelessWidget
    0.46
     StatefulWidget
    0.39
    Big
    0.35
     Frankrijk
    0.33
     Has
    0.33
     Big
    0.33
     hvordan
    0.31
     przeznac
    0.30
     oorsp
    0.30
    ]?.
    0.30
    Activations Density 0.015%

    No Known Activations