© 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
    3. Qwen2.5-7B-IT
    5. 20-MATRYOSHKA-65K
    7. 2074
    Prev
    Next
    INDEX
    Explanations

    code

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    chanind/qwen2.5-7B-it-layer-20-saes/pile/matryoshka/k-100
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    #aa
    -0.30
    :numel
    -0.26
    )NULL
    -0.24
    #af
    -0.24
    :maj
    -0.23
    #ac
    -0.23
    :animated
    -0.22
    :');↵
    -0.21
    .ExecuteReader
    -0.21
    >manual
    -0.20
    POSITIVE LOGITS
    /respond
    0.23
     withStyles
    0.22
     cocks
    0.21
     }])↵
    0.20
    éķ§
    0.20
     tabIndex
    0.19
     nudity
    0.19
    è£ħ
    0.19
     ActionTypes
    0.19
     pione
    0.18
    Activations Density 0.001%

    No Known Activations