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-2-2B
    4. Residual Stream - 16k
    5. 15-GEMMASCOPE-RES-16K
    6. 1532
    Prev
    Next
    INDEX
    Explanations

    phrases indicating comparisons, choices, or allegations linked to political figures or situations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.15.hook_resid_post
    Hook Layer
    15
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    tvguidetime
    -0.96
     Efq
    -0.93
     purpoſe
    -0.85
     uſed
    -0.80
     myſelf
    -0.77
     Jefus
    -0.76
     ſta
    -0.75
     poffe
    -0.75
     fufficient
    -0.74
     fevere
    -0.73
    POSITIVE LOGITS
     mergeFrom
    0.88
     '\\;'
    0.79
    IsContent
    0.76
    Obrázky
    0.67
     ModelExpression
    0.66
    AnchorStyles
    0.65
     nahilalakip
    0.63
     Réponses
    0.62
    исленность
    0.60
    AndEndTag
    0.59
    Activations Density 0.149%

    No Known Activations