© 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. 4241
    Prev
    Next
    INDEX
    Explanations

    pieces of text discussing logical reasoning and systems

    oai_token-act-pair · gpt-3.5-turboTriggered 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
    487
    +0.10
    0.5%
    642
    +0.10
    0.5%
    635
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    635
    +0.10
    0.01
    255
    +0.10
    0.01
    613
    +0.10
    0.01
    Negative Logits
    lain
    -0.71
     Volunte
    -0.68
     Leopard
    -0.68
    avez
    -0.66
    orks
    -0.66
    eded
    -0.65
    meet
    -0.65
    hold
    -0.63
    artney
    -0.62
    rifice
    -0.61
    POSITIVE LOGITS
     logic
    0.92
     reasoning
    0.91
     fallacy
    0.91
    matical
    0.89
     underpin
    0.87
    istically
    0.86
    DragonMagazine
    0.81
     puzzles
    0.81
    istical
    0.79
     posit
    0.78
    Activations Density 0.022%

    No Known Activations