© 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. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    5. GPT2-Small
    7. Residual Similar CE E2E Recon
    9. 6-RES_SCEFR-AJT
    11. 1090
    Prev
    Next
    INDEX
    Explanations

    statements where individuals claim to possess or believe in certain things

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    neuronpedia/gpt2-small__res_scefr-ajt/6-res_scefr-ajt
    Prompts (Dashboard)
    12,288 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    46,080
    Data Type
    torch.float32
    Hook Point
    blocks.6.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    apollo-research/Skylion007-openwebtext-tokenizer-gpt2
    Hook Point Layer
    6
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    411
    +0.14
    0.5%
    393
    +0.14
    0.5%
    92
    +0.12
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    566
    +0.14
    0.07
    528
    +0.14
    0.06
    762
    +0.12
    0.09
    Negative Logits
    arnaev
    -0.93
     Yards
    -0.86
    inson
    -0.85
    ousand
    -0.84
    utherland
    -0.84
    terness
    -0.83
    ighed
    -0.83
    onential
    -0.82
    Side
    -0.81
    osion
    -0.81
    POSITIVE LOGITS
     invincible
    1.07
     kindred
    0.93
    ë
    0.87
     innocence
    0.87
     constitu
    0.86
     protecting
    0.86
     benevolent
    0.85
     fide
    0.85
     unbeat
    0.84
     paranormal
    0.84
    Activations Density 1.557%

    No Known Activations