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

    Neuronpedia

    Jacobian LensNEW
    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. 20554
    Prev
    Next
    INDEX
    Explanations

    controversial topics or issues

    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
    288
    +0.17
    0.8%
    447
    +0.14
    0.7%
    151
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    151
    +0.17
    0.03
    288
    +0.14
    0.02
    408
    +0.12
    0.03
    Negative Logits
    á
    -0.86
    elsen
    -0.84
    vation
    -0.81
    abetic
    -0.75
    ©¶æ
    -0.74
    ILA
    -0.69
    ĺħ
    -0.67
    ynthesis
    -0.66
    ruary
    -0.66
    abetes
    -0.65
    POSITIVE LOGITS
    ity
    0.86
     aspects
    0.85
     topics
    0.85
     controversial
    0.81
     fringe
    0.78
     topic
    0.77
     decisions
    0.75
    iating
    0.73
     tactics
    0.73
     opinions
    0.73
    Activations Density 0.052%

    No Known Activations