© 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
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 37-GEMMASCOPE-RES-16K
    6. 4181
    Prev
    Next
    INDEX
    Explanations

    specific time formats and timestamps

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 37-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_37/width_16k/average_l0_124
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.37.hook_resid_post
    Hook Layer
    37
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    午後
    -0.54
     препратки
    -0.54
     CanadaChoose
    -0.54
     CURIAM
    -0.47
     вечером
    -0.46
     dusk
    -0.44
     pm
    -0.43
     Evening
    -0.43
    存于互联网档案馆
    -0.42
    پیوند
    -0.42
    POSITIVE LOGITS
     AM
    0.77
    am
    0.70
    AM
    0.69
     am
    0.69
     morning
    0.67
     TextAppearance
    0.56
    morning
    0.51
    Am
    0.50
     Am
    0.50
     Amm
    0.48
    Activations Density 0.133%

    No Known Activations