© 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. 11957
    Prev
    Next
    INDEX
    Explanations

    names of specific individuals, likely related to politics or public figures

    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
    481
    +0.17
    0.8%
    608
    +0.14
    0.7%
    745
    +0.14
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    15
    +0.17
    0.07
    362
    +0.14
    0.07
    608
    +0.14
    0.06
    Negative Logits
    utenant
    -0.76
    olin
    -0.74
    holes
    -0.73
    lehem
    -0.73
    lator
    -0.71
    anguage
    -0.70
    glomer
    -0.68
    virtual
    -0.67
    interstitial
    -0.66
    stress
    -0.66
    POSITIVE LOGITS
     refusal
    1.35
     insistence
    1.32
     efforts
    1.28
     plight
    1.25
     inability
    1.23
     actions
    1.22
     unwillingness
    1.21
     assertion
    1.21
     antics
    1.20
     involvement
    1.19
    Activations Density 0.234%

    No Known Activations