© 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. 2413
    Prev
    Next
    INDEX
    Explanations

    mentions where someone is blocked or unblocked in a digital context

    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
    585
    +0.13
    0.6%
    9
    +0.11
    0.5%
    403
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    585
    +0.13
    0.02
    740
    +0.11
    0.02
    557
    +0.11
    0.01
    Negative Logits
    matically
    -0.89
    marked
    -0.78
    shown
    -0.73
    tis
    -0.72
    negie
    -0.71
    present
    -0.68
    resent
    -0.68
    pled
    -0.67
    adra
    -0.66
    verty
    -0.66
    POSITIVE LOGITS
     Susp
    0.85
    âĦ¢:
    0.83
     Away
    0.83
    adoes
    0.82
     Favorite
    0.80
     Accounts
    0.79
     Removed
    0.78
     Hearing
    0.78
     List
    0.75
     Cust
    0.74
    Activations Density 0.049%

    No Known Activations