Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Gemma-2-2B
    3. 7-GEMMASCOPE-TRANSCODER-16K
    4. 6963
    Prev
    Next
    INDEX
    Explanations

    words related to deception, both in a game context and in general.

    oai_token-act-pair · gemini-2.0-flash

    attends to terms related to user engagement from terms related to technology functions or challenges.

    oai_attention-head · gpt-4o-miniTriggered by @mihirpanchal5400

    concealment, disguise

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_7/width_16k/average_l0_70
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.7.ln2.hook_normalized
    Architecture
    jumprelu_transcoder
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     fake
    -0.66
     mislead
    -0.60
     Fake
    -0.58
     faked
    -0.58
    fake
    -0.57
     dummy
    -0.56
     illusion
    -0.56
     misleading
    -0.52
     pretending
    -0.50
    Fake
    -0.48
    POSITIVE LOGITS
    ArgsConstructor
    0.70
     <>",
    0.69
    Bakgrunnsstoff
    0.66
     automatiques
    0.65
    ніципалі
    0.63
    ViewFeatures
    0.62
     Marginal
    0.60
     aanges
    0.60
     nakalista
    0.60
    новниш
    0.59
    Activations Density 19.835%

    No Known Activations