© 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. 20-GEMMASCOPE-RES-16K
    6. 2155
    Prev
    Next
    INDEX
    Explanations

    strong expressions of personal sentiment or emotion

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

    first-person pronouns like "I" and "I'm".

    oai_token-act-pair · claude-3-5-haiku-20241022Triggered by @neilrathi
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    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
     itſelf
    -0.49
     Bean
    -0.45
     setId
    -0.44
     alpina
    -0.44
    󠁿
    -0.43
    ARANCE
    -0.43
     noqa
    -0.41
    JAR
    -0.41
     inputStream
    -0.41
    Bean
    -0.40
    POSITIVE LOGITS
     aint
    0.68
     wasnt
    0.67
    itsubishi
    0.66
    wiliwch
    0.65
     didnt
    0.64
     Paglinawan
    0.59
     outta
    0.58
     gonna
    0.58
     couldnt
    0.58
    ientras
    0.58
    Activations Density 0.064%

    No Known Activations