© 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
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 9611
    Prev
    Next
    INDEX
    Explanations

    nice followed by positive descriptors or 'and'

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

    The neuron is keyed to positive evaluative adjectives (especially “nice,” “good,” “lovely,” etc.) indicating favorable sentiment.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    the word "nice" in positive or descriptive contexts.

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

    a mild, casual positive adjective expressing approval (often in “adj and …” constructions)

    oai_token-act-pair · gpt-5Triggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_24_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     cinética
    1.23
    ième
    1.15
    nk
    1.13
     pensé
    1.13
    ます
    1.10
     científica
    1.09
    s
    1.09
    Η
    1.09
    opcao
    1.09
    nent
    1.08
    POSITIVE LOGITS
    ა
    1.16
     specular
    1.08
     tingling
    1.08
    тар
    1.07
     orderly
    1.04
    ток
    1.01
     berth
    0.95
    าว
    0.93
     plump
    0.93
     forgery
    0.93
    Activations Density 0.018%

    No Known Activations