© 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. 8835
    Prev
    Next
    INDEX
    Explanations

    actions related to reviewing, investigating, or interviewing

    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
    715
    +0.12
    0.6%
    654
    +0.12
    0.5%
    161
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    654
    +0.12
    0.03
    405
    +0.12
    0.04
    715
    +0.11
    0.03
    Negative Logits
    guard
    -0.77
    Torrent
    -0.75
    bott
    -0.68
    cil
    -0.67
    trap
    -0.66
    vic
    -0.66
    arak
    -0.66
    emb
    -0.64
    âĹ¼
    -0.64
    lez
    -0.64
    POSITIVE LOGITS
    ively
    0.89
     favorably
    0.82
     feasibility
    0.78
     carefully
    0.74
     them
    0.74
     whether
    0.73
     how
    0.72
    matically
    0.72
     trends
    0.71
     incoming
    0.70
    Activations Density 0.127%

    No Known Activations