© 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. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 24576
    9. 8-RES_FS24576-JB
    11. 24499
    Prev
    Next
    INDEX
    Explanations

    mentions of "Zeppelin."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    57
    +0.13
    0.6%
    555
    +0.12
    0.6%
    670
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    57
    +0.13
    0.01
    555
    +0.12
    0.01
    670
    +0.12
    0.01
    Negative Logits
    DOWN
    -0.71
    Interstitial
    -0.67
    ainers
    -0.67
    wear
    -0.64
     provisional
    -0.64
    channelAvailability
    -0.63
    aneously
    -0.63
    CLASSIFIED
    -0.63
    ials
    -0.63
    aneous
    -0.63
    POSITIVE LOGITS
    ppelin
    1.37
    lda
    1.25
    ppel
    1.16
    pez
    1.04
    plin
    1.00
    iger
    0.95
    itsch
    0.94
    ichen
    0.93
    lder
    0.89
    ven
    0.89
    Activations Density 0.015%

    No Known Activations