© 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 12288
    9. 8-RES_FS12288-JB
    11. 1831
    Prev
    Next
    INDEX
    Explanations

    words related to communication or storytelling

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    597
    +0.10
    0.5%
    96
    +0.10
    0.4%
    184
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    184
    +0.10
    0.03
    597
    +0.10
    0.04
    213
    +0.09
    0.03
    Negative Logits
    imposed
    -0.71
    adesh
    -0.66
    ccording
    -0.65
     sidx
    -0.64
    cells
    -0.64
    berus
    -0.62
    zinski
    -0.61
    rane
    -0.60
    urdue
    -0.60
    nam
    -0.59
    POSITIVE LOGITS
    tale
    1.53
    ingly
    1.06
     us
    1.03
     me
    0.89
     tales
    0.89
     tale
    0.87
     him
    0.84
     stories
    0.83
     lies
    0.79
     Tale
    0.78
    Activations Density 0.082%

    No Known Activations