© 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. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    3. GPT2-Small
    4. Residual Similar CE E2E Recon
    5. 6-RES_SCEFR-AJT
    6. 2498
    Prev
    Next
    INDEX
    Explanations

    religious and spiritual references, including prayers and religious practices

    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
    70
    +0.16
    0.6%
    288
    +0.15
    0.6%
    468
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    70
    +0.16
    0.15
    665
    +0.15
    0.13
    468
    +0.12
    0.09
    Negative Logits
    DragonMagazine
    -1.06
    fal
    -0.86
    of
    -0.83
    Of
    -0.82
    ãĥ¼ãĥ³
    -0.79
    OF
    -0.78
    ãĥ¼ãĥ«
    -0.77
    GV
    -0.77
    igers
    -0.76
    drawn
    -0.75
    POSITIVE LOGITS
     furiously
    1.16
     extensively
    1.14
     differently
    1.04
     passionately
    1.00
     against
    0.98
     loudly
    0.97
     outdoors
    0.95
     tirelessly
    0.94
     vigorously
    0.93
     aloud
    0.93
    Activations Density 2.164%

    No Known Activations