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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B
    7. Residual Stream - 131k
    9. 20-GEMMASCOPE-RES-131K
    11. 105677
    Prev
    Next
    INDEX
    Explanations

    emotional expressions and sentiments in written feedback

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

    Connecting words or phrases that link narrative elements, often used in explanatory contexts or to establish relationships between described people, events, or concepts.

    eleuther_acts_top20 · claude-3-7-sonnet-20250219Triggered by @lu-christina
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114
    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
    AccessorTable
    -0.64
    setVerticalGroup
    -0.60
    Autoritní
    -0.55
    GOTREF
    -0.53
    wäh
    -0.50
     CYTHON
    -0.50
     Muda
    -0.49
    RTEE
    -0.49
     Sle
    -0.48
     SMT
    -0.48
    POSITIVE LOGITS
     solidaridad
    0.39
     sentimental
    0.37
     compassion
    0.36
     spiritually
    0.36
    rawDesc
    0.35
     charitable
    0.35
     mournful
    0.35
     punggung
    0.35
     charity
    0.34
     ProtoMessage
    0.34
    Activations Density 0.078%

    No Known Activations