© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 10401
    Prev
    Next
    INDEX
    Explanations

    numerical information related to levels and scores

    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
    Neuron Alignment
    Index
    Value
    % of L₁
    99
    +0.15
    0.5%
    647
    +0.15
    0.5%
    466
    +0.13
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    647
    +0.15
    0.04
    198
    +0.15
    0.04
    99
    +0.13
    0.05
    Negative Logits
    vous
    -0.99
    ILLE
    -0.98
     Andromeda
    -0.97
    fect
    -0.95
    oute
    -0.90
    uter
    -0.89
    enegger
    -0.86
    è¦
    -0.86
    lez
    -0.85
     Arg
    -0.84
    POSITIVE LOGITS
     crossings
    1.08
    level
    0.93
     rise
    0.92
     elevation
    0.91
    xual
    0.90
     assurance
    0.89
     levels
    0.85
    ropri
    0.85
     attainment
    0.84
    eways
    0.84
    Activations Density 0.628%

    No Known Activations