© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B
    7. Residual Stream - 16k
    9. 31-GEMMASCOPE-RES-16K
    11. 4097
    Prev
    Next
    INDEX
    Explanations

    specific technical terms and configurations related to programming or system functions

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

    hexadecimal numbers and system constants in source code.

    oai_token-act-pair · claude-4-5-haikuTriggered by @emiglarou

    hexadecimal constant definitions and flag values in system programming code.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    code fragments from low-level systems programming that manipulate or print bitmask flags and options, featuring all-caps underscore constants and hex values.

    oai_token-act-pair · gpt-5Triggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-9B @ 31-gemmascope-res-16k
    Configuration
    google/gemma-scope-9b-pt-res/layer_31/width_16k/average_l0_114
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.31.hook_resid_post
    Hook Layer
    31
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    principalColumn
    -0.52
     ModelExpression
    -0.50
     Infórmanos
    -0.47
    Personendaten
    -0.47
     Infór
    -0.46
    exitRule
    -0.45
    StoryboardSegue
    -0.45
    isContained
    -0.45
     Trung
    -0.44
    adaptiveStyles
    -0.44
    POSITIVE LOGITS
     flags
    1.03
     Flags
    0.98
     bit
    0.96
    flags
    0.93
    Flags
    0.92
     flag
    0.85
     bits
    0.84
     Flag
    0.84
    Flag
    0.81
    bit
    0.78
    Activations Density 1.029%

    No Known Activations