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

    Neuronpedia

    Jacobian LensNEW
    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. 14984
    Prev
    Next
    INDEX
    Explanations

    phrases related to accidental actions or mistakes

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

    accidental or unintended actions and their consequences.

    oai_token-act-pair · claude-4-5-haikuTriggered by @emiglarou
    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
    OGND
    -0.61
    tagHelperRunner
    -0.59
    awtextra
    -0.57
     protoimpl
    -0.54
    WebMethod
    -0.53
    PreExecute
    -0.48
    testify
    -0.46
    +:+
    -0.46
    WriteBarrier
    -0.45
     nahilalakip
    -0.45
    POSITIVE LOGITS
     accidentally
    2.28
     inadvertently
    2.11
     unintentionally
    1.88
     accidental
    1.86
     inadvert
    1.85
     mistakenly
    1.82
     unintentional
    1.63
     unwittingly
    1.60
     unknowingly
    1.54
     Accidental
    1.51
    Activations Density 0.861%

    No Known Activations