© 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. 867
    Prev
    Next
    INDEX
    Explanations

    contact information and communication-related terms (e.g., email, Twitter)

    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
    284
    +0.15
    0.5%
    50
    +0.15
    0.5%
    737
    +0.14
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    50
    +0.15
    0.03
    679
    +0.15
    0.04
    61
    +0.14
    0.02
    Negative Logits
     gou
    -0.94
     bulldo
    -0.90
     imported
    -0.84
     furn
    -0.84
     1893
    -0.83
     importing
    -0.83
     onwards
    -0.81
     cart
    -0.81
    Christmas
    -0.80
     ho
    -0.80
    POSITIVE LOGITS
    ioned
    1.21
    ing
    0.97
    ed
    0.97
    linger
    0.96
    ting
    0.95
    fort
    0.95
    iquette
    0.94
    patrick
    0.93
    fax
    0.93
    Wire
    0.91
    Activations Density 0.220%

    No Known Activations