© 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. Gemma-3-27B-IT
    5. 40-GEMMASCOPE-2-RES-262K
    7. 4353
    Prev
    Next
    INDEX
    Explanations

    instructions and queries

    np_acts-logits-general · gemini-2.5-flash-lite

    text that contains explicit instructions, rules, or constraints directing the assistant's behavior (system prompts and policy-style directives).

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0

    language that conveys formal task specifications—constraints, procedural instructions, policies, links/resources, templates/formats, and feature requirements.

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/resid_post/layer_40_width_262k_l0_medium
    Prompts (Dashboard)
    238,145 prompts, 512 tokens each
    Dataset (Dashboard)
    lmsys + oasst1
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Incidentally
    0.41
    defn
    0.40
     doubtless
    0.39
     paltry
    0.38
    <unused303>
    0.38
     ostensibly
    0.38
     stalwart
    0.37
    <unused2049>
    0.36
    <unused267>
    0.36
    THRESH
    0.36
    POSITIVE LOGITS
     bellow
    0.61
    ´
    0.60
     advices
    0.56
     planification
    0.56
     lenght
    0.56
     ressources
    0.55
     Nowadays
    0.55
     wich
    0.53
     partecip
    0.52
     restauration
    0.52
    Activations Density 0.073%

    No Known Activations