© 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 12288
    9. 8-RES_FS12288-JB
    11. 9752
    Prev
    Next
    INDEX
    Explanations

    details about different parts or sections in a location or a written piece

    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
    216
    +0.11
    0.5%
    659
    +0.10
    0.5%
    344
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    216
    +0.11
    0.04
    597
    +0.10
    0.03
    344
    +0.09
    0.03
    Negative Logits
    culosis
    -0.70
    gently
    -0.70
    berus
    -0.70
     Skies
    -0.64
    ministic
    -0.62
     Arms
    -0.61
     Klux
    -0.59
    rily
    -0.59
     appropri
    -0.59
    ãĤī
    -0.59
    POSITIVE LOGITS
    ridges
    1.09
    ridge
    1.08
     thereof
    0.89
    meal
    0.88
    ials
    0.88
    icular
    0.83
     of
    0.81
    atum
    0.77
    icle
    0.76
    aking
    0.75
    Activations Density 0.074%

    No Known Activations