© 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. 36-GEMMASCOPE-RES-16K
    6. 15842
    Prev
    Next
    INDEX
    Explanations

    technical issues and programming-related errors

    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
     '\\;'
    -0.49
    Népesség
    -0.46
    AndEndTag
    -0.40
    oredCriteria
    -0.40
    onViewCreated
    -0.37
    setFocus
    -0.36
     Vikipedi
    -0.36
    keyColumn
    -0.35
    concentr
    -0.35
     concentr
    -0.35
    POSITIVE LOGITS
     workaround
    1.20
     trick
    0.81
     hack
    0.74
     circumvent
    0.73
     tricks
    0.72
     loophole
    0.72
     undocumented
    0.70
     clever
    0.69
     hacks
    0.69
     tricky
    0.68
    Activations Density 1.264%

    No Known Activations