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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B
    4. Residual Stream - 16k
    5. 38-GEMMASCOPE-RES-16K
    6. 10255
    Prev
    Next
    INDEX
    Explanations

    terms related to gauge fixing and transformations in a mathematical context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 38-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_38/width_16k/average_l0_128
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.38.hook_resid_post
    Hook Layer
    38
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     '{@
    -0.45
     mérit
    -0.42
    
    -0.41
     Krä
    -0.38
    CONFIRM
    -0.37
     contribu
    -0.36
    一目
    -0.36
     illustrating
    -0.36
     confirma
    -0.35
     celebrar
    -0.35
    POSITIVE LOGITS
     reform
    0.60
    rrggbb
    0.57
     impose
    0.57
    PreExecute
    0.55
    Gön
    0.53
     treat
    0.53
    الدراسه
    0.52
     recast
    0.52
     vary
    0.51
     DBNull
    0.51
    Activations Density 1.788%

    No Known Activations