© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-3-27B-IT
    3. 14-GEMMASCOPE-2-TRANSCODER-262K
    4. 97239
    Prev
    Next
    INDEX
    Explanations

    The provided lists offer clues to the neuron's behavior. Let's break them down:1. **MAX_ACTIVATING_TOKENS**: This list includes words like 'Lear', 'antil', 'planted', 'friend', 'story', 'think'.2. **TOKENS_AFTER_MAX_ACTIVATING_TOKEN**: * 'Lear' is followed by 'ner' -> "Learner" * 'antil' is followed by 'ism' -> "antilism" (likely referring to systems or ideologies) * 'planted' is followed by 'in' -> "planted in" (indicating location or context) * 'friend' is followed by 'involving' -> "friend involving" (suggesting relationships or situations) * 'story' is followed by 'about' -> "story about"3. **TOP_POSITIVE_LOGITS**: This list contains a mix of languages and entities: 'Tell', 'Humans', 'Mary

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/transcoder_all/layer_14_width_262k_l0_small_affine
    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
    vn
    0.68
     ring
    0.56
    ing
    0.54
    onder
    0.53
    indle
    0.53
    v
    0.53
     clash
    0.52
    iggins
    0.52
    ti
    0.51
    ths
    0.51
    POSITIVE LOGITS
    Tell
    0.59
    Humans
    0.55
    ਼
    0.54
    ོན་
    0.53
    Mary
    0.51
    的
    0.50
    (
    0.49
    Deux
    0.48
     températures
    0.48
    Mare
    0.48
    Activations Density 0.000%

    No Known Activations