© 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
    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. 11895
    Prev
    Next
    INDEX
    Explanations

    patterns of numbers and special characters, such as specific sequences and digits

    oai_token-act-pair · gpt-3.5-turbo

    use of special characters or symbols, specifically variations of 'Ģ', 'ľ', 'Ŀ', and 'Ļ', within the text

    oai_token-act-pair · gpt-3.5-turboTriggered by @danbraun
    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
    160
    +0.12
    0.5%
    327
    +0.11
    0.5%
    577
    +0.11
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    338
    +0.12
    0.22
    731
    +0.11
    0.18
    48
    +0.11
    0.21
    Negative Logits
     Palestin
    -0.80
    geries
    -0.77
     scram
    -0.75
     Guatem
    -0.71
     Chattanooga
    -0.70
     MV
    -0.69
     convenience
    -0.68
     incorpor
    -0.68
     sacrific
    -0.68
     elig
    -0.67
    POSITIVE LOGITS
    ï¸ı
    1.13
    audi
    0.97
    £
    0.97
    âĶĢâĶĢ
    0.95
    else
    0.90
    âĶĢâĶĢâĶĢâĶĢ
    0.88
    endif
    0.88
    wise
    0.85
    `
    0.84
    ¯¯
    0.84
    Activations Density 2.264%

    No Known Activations