© 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
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-9B
    7. Residual Stream - 16k
    9. 31-GEMMASCOPE-RES-16K
    11. 16153
    Prev
    Next
    INDEX
    Explanations

    phrases indicating something has gone wrong or problems occurring

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

    phrases expressing that something has failed or gone wrong.

    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
    exitRule
    -0.41
     colis
    -0.38
    abetes
    -0.36
    cplusplus
    -0.35
     bestaan
    -0.35
     scio
    -0.34
    BarStyle
    -0.33
     Konferenz
    -0.33
     perto
    -0.32
     الرياضيه
    -0.31
    POSITIVE LOGITS
     wrong
    1.74
    wrong
    1.53
     Wrong
    1.41
     WRONG
    1.38
    Wrong
    1.31
    WRONG
    1.08
     wrongs
    1.01
     amiss
    0.94
     errado
    0.77
     incorrect
    0.77
    Activations Density 0.183%

    No Known Activations