© 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
    4. Residual Stream - 16k
    5. 31-GEMMASCOPE-RES-16K
    6. 15714
    Prev
    Next
    INDEX
    Explanations

    characters and their interactions in narrative contexts

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

    narrative sequences describing events or steps that unfold in chronological order.

    oai_token-act-pair · claude-4-5-haikuTriggered by @emiglarou
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 31-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_31/width_16k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     consisted
    -0.54
     weren
    -0.50
     löyty
    -0.46
    bahaya
    -0.45
     existed
    -0.45
     belonged
    -0.44
     bestaan
    -0.44
     belong
    -0.43
     appartiennent
    -0.43
     resembled
    -0.42
    POSITIVE LOGITS
     gets
    1.29
     takes
    1.21
     goes
    1.18
     becomes
    1.15
     receives
    1.03
     develops
    1.03
     makes
    1.02
     begins
    1.02
     starts
    1.00
     decides
    1.00
    Activations Density 2.660%

    No Known Activations