© 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. 1027
    Prev
    Next
    INDEX
    Explanations

    responses indicating a direct personal experience or perspective

    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
    661
    +0.13
    0.4%
    718
    +0.13
    0.4%
    168
    +0.12
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    159
    +0.13
    0.05
    589
    +0.13
    0.05
    105
    +0.12
    0.04
    Negative Logits
    ibaba
    -1.15
    hillary
    -1.15
    artments
    -1.00
    inge
    -1.00
    netflix
    -1.00
    selage
    -0.98
    eeper
    -0.97
    erva
    -0.97
    Thumbnail
    -0.96
    dos
    -0.95
    POSITIVE LOGITS
     damned
    0.93
     uncond
    0.93
     Clive
    0.88
     Pers
    0.84
     overr
    0.84
     Hercules
    0.82
     Bers
    0.82
     Cann
    0.82
     Carth
    0.81
     Lucky
    0.81
    Activations Density 1.343%

    No Known Activations