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
    4. Residual Stream - 16k
    5. 21-GEMMASCOPE-RES-16K
    6. 14944
    Prev
    Next
    INDEX
    Explanations

    actions and emotional responses in a narrative context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    WithIOException
    -0.57
    RRect
    -0.53
    $_['
    -0.49
     ось
    -0.49
    makeConstraints
    -0.49
     ROYAL
    -0.48
     législation
    -0.47
    juris
    -0.47
    wiſe
    -0.47
     harvest
    -0.47
    POSITIVE LOGITS
    ScopeManager
    0.51
    SequentialGroup
    0.42
     slid
    0.38
    Personendaten
    0.36
    RTHOOK
    0.35
     sliding
    0.34
     standing
    0.33
     frowned
    0.32
     pareció
    0.32
     motionless
    0.32
    Activations Density 0.131%

    No Known Activations