© 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. 2642
    Prev
    Next
    INDEX
    Explanations

    passages related to physical actions or experiences, particularly those involving an external physical event or condition

    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
    238
    +0.17
    0.6%
    288
    +0.15
    0.5%
    59
    +0.15
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.17
    0.10
    238
    +0.15
    0.06
    326
    +0.15
    0.09
    Negative Logits
    ctions
    -0.91
    itars
    -0.81
    orio
    -0.80
    NOR
    -0.78
    erent
    -0.78
     Sabha
    -0.77
    rade
    -0.76
    Refer
    -0.76
    integer
    -0.75
     recomm
    -0.73
    POSITIVE LOGITS
    alls
    1.01
     blanket
    0.96
     veil
    0.94
     Veil
    0.89
    Versions
    0.86
     covering
    0.86
    umin
    0.85
     covers
    0.84
    lasses
    0.82
    ãĤ¹
    0.82
    Activations Density 1.462%

    No Known Activations