© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 20333
    Prev
    Next
    INDEX
    Explanations

    words related to desire or preference

    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
    38
    +0.12
    0.6%
    459
    +0.11
    0.5%
    750
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    431
    +0.12
    0.05
    38
    +0.11
    0.05
    566
    +0.11
    0.05
    Negative Logits
    ulty
    -0.67
    hiba
    -0.66
    semble
    -0.65
    ilings
    -0.65
    igmatic
    -0.64
    soDeliveryDate
    -0.63
    ccording
    -0.62
    unny
    -0.61
    rup
    -0.61
    livious
    -0.61
    POSITIVE LOGITS
     revenge
    0.86
    reprene
    0.84
     to
    0.81
     nothing
    0.80
    only
    0.79
     something
    0.76
     answers
    0.74
    lessly
    0.71
    WARD
    0.68
     desperately
    0.67
    Activations Density 0.080%

    No Known Activations