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. 2547
    Prev
    Next
    INDEX
    Explanations

    references to license agreements or legal terms related to software usage

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

    code comments

    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
    migrationBuilder
    -0.34
    SerializedName
    -0.31
    appBar
    -0.31
    phosa
    -0.29
     DialogResult
    -0.29
    isFile
    -0.29
     divisors
    -0.28
     revisions
    -0.28
     رق
    -0.28
    SuspendLayout
    -0.28
    POSITIVE LOGITS
    httphttps
    0.82
     désolés
    0.69
    																									
    0.65
    ————————————
    0.65
    ————————————————
    0.65
    																													
    0.64
    pingente
    0.63
     corações
    0.63
                                   
    0.62
    Архівовано
    0.62
    Activations Density 0.203%

    No Known Activations