© 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. 2699
    Prev
    Next
    INDEX
    Explanations

    references to a specific organization, likely related to news or activism

    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
    496
    +0.16
    0.7%
    336
    +0.14
    0.7%
    161
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    496
    +0.16
    0.02
    247
    +0.14
    0.01
    733
    +0.12
    0.01
    Negative Logits
    ensions
    -0.70
    utic
    -0.67
    omething
    -0.66
    cone
    -0.65
    ön
    -0.65
     valve
    -0.63
    lycer
    -0.62
     pastry
    -0.62
    enhagen
    -0.62
     date
    -0.61
    POSITIVE LOGITS
    uggets
    0.94
    arium
    0.92
    ally
    0.88
     Travels
    0.87
    icans
    0.82
    wide
    0.80
    naire
    0.80
    Wars
    0.78
    ican
    0.77
    ukong
    0.75
    Activations Density 0.014%

    No Known Activations