© 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. 11983
    Prev
    Next
    INDEX
    Explanations

    words related to the hexagonal shape

    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
    553
    +0.13
    0.6%
    651
    +0.13
    0.6%
    516
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    284
    +0.13
    0.02
    651
    +0.13
    0.02
    289
    +0.12
    0.01
    Negative Logits
    jriwal
    -0.77
    Interstitial
    -0.75
     Resort
    -0.69
    utenant
    -0.69
    getic
    -0.67
     Loan
    -0.67
     Majority
    -0.65
     conditioning
    -0.65
     cessation
    -0.63
     Provider
    -0.63
    POSITIVE LOGITS
    agonal
    1.44
    agon
    1.12
    cellent
    1.04
    agons
    0.93
    es
    0.91
    agram
    0.88
    actly
    0.87
    aminer
    0.87
    dump
    0.85
    iates
    0.82
    Activations Density 0.032%

    No Known Activations