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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 19801
    Prev
    Next
    INDEX
    Explanations

    references to physical locations, specifically basements

    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
    447
    +0.13
    0.6%
    481
    +0.13
    0.6%
    288
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.13
    0.03
    326
    +0.13
    0.02
    40
    +0.12
    0.02
    Negative Logits
    andum
    -0.89
    nir
    -0.87
    uality
    -0.86
    anchester
    -0.77
    utan
    -0.77
    wana
    -0.74
    ivia
    -0.71
    onen
    -0.69
    aim
    -0.69
    oulos
    -0.69
    POSITIVE LOGITS
     basement
    1.09
     attic
    1.07
     closet
    1.06
     door
    1.03
     fireplace
    1.02
     lid
    0.97
     renovation
    0.95
     doors
    0.94
     apartment
    0.94
     cellar
    0.91
    Activations Density 0.056%

    No Known Activations