Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 36-GEMMASCOPE-RES-16K
    6. 11429
    Prev
    Next
    INDEX
    Explanations

    terms related to semi-structured or semi-parametric models and structures

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 36-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_36/width_16k/average_l0_120
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.36.hook_resid_post
    Hook Layer
    36
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     innov
    -0.47
    foul
    -0.46
    日閲覧
    -0.45
     Winaray
    -0.45
     innovate
    -0.44
    TRIBUN
    -0.44
    actéristi
    -0.44
     Chwiliwch
    -0.43
     INNOV
    -0.43
    gameserver
    -0.43
    POSITIVE LOGITS
     Semi
    0.97
     semi
    0.93
     sem
    0.90
     Sem
    0.87
    Sem
    0.85
     SEMI
    0.81
    Semi
    0.78
     SEM
    0.71
     semis
    0.66
    SEM
    0.63
    Activations Density 0.205%

    No Known Activations