© 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 - 16k
    9. 20-GEMMASCOPE-RES-16K
    11. 2682
    Prev
    Next
    INDEX
    Explanations

    conjunctions and negations in complex sentence structures

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

    legal language and modal verbs expressing possibility or obligation (like 'should', 'could', 'might', 'must').

    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
     
    -0.43
    .
    -0.41
    T
    -0.35
     which
    -0.34
     fordi
    -0.33
    -
    -0.32
     The
    -0.32
     based
    -0.31
    組み
    -0.31
     Orozco
    -0.31
    POSITIVE LOGITS
    ftagPool
    0.88
    ſelves
    0.69
    ſelf
    0.68
    ロウィン
    0.67
     reaſon
    0.66
    ſchaft
    0.66
    нгред
    0.66
    linawan
    0.66
    BibitemShut
    0.65
     Reſ
    0.65
    Activations Density 0.312%

    No Known Activations