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. 46108
    Prev
    Next
    INDEX
    Explanations

    technical terms and programming-related concepts

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

    registration, references, nominate, browse

    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
     دام
    -0.38
    jag
    -0.36
    Horizon
    -0.34
     Jag
    -0.33
    Jag
    -0.33
     lant
    -0.32
     jag
    -0.29
    pern
    -0.29
     Horizon
    -0.29
    Delta
    -0.29
    POSITIVE LOGITS
    OrBuilder
    0.64
    сылкі
    0.59
     betweenstory
    0.58
    oredCriteria
    0.54
    ArrowToggle
    0.52
     ProtoMessage
    0.52
    parsedMessage
    0.50
    StoreMessageInfo
    0.49
    ########.
    0.48
    MemoryWarning
    0.47
    Activations Density 0.395%

    No Known Activations