© 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. Joseph Bloom · Feature Splitting for GPT2-Small
    5. GPT2-Small
    7. Residual Feature Splitting 24576
    9. 8-RES_FS24576-JB
    11. 10436
    Prev
    Next
    INDEX
    Explanations

    instances where someone is discussing or backing up assertions with additional information

    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
    125
    +0.09
    0.4%
    713
    +0.09
    0.4%
    340
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    713
    +0.09
    0.01
    125
    +0.09
    0.01
    747
    +0.09
    0.01
    Negative Logits
    avery
    -0.74
    eteenth
    -0.68
    omal
    -0.68
    OY
    -0.67
    jee
    -0.65
    ities
    -0.62
    asting
    -0.62
    acht
    -0.62
    ====
    -0.61
    theless
    -0.61
    POSITIVE LOGITS
    stretched
    0.79
     against
    0.79
     shaky
    0.77
    packs
    0.72
     Rept
    0.72
     Arkham
    0.69
    stock
    0.68
    puff
    0.68
    Tes
    0.67
    graded
    0.67
    Activations Density 0.022%

    No Known Activations