© 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 EvalsExportsinterp-engineNEWAPI 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. 15732
    Prev
    Next
    INDEX
    Explanations

    verbs indicating uncertainty or possibility

    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
    559
    +0.10
    0.5%
    8
    +0.10
    0.5%
    71
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    551
    +0.10
    0.07
    8
    +0.10
    0.06
    49
    +0.09
    0.06
    Negative Logits
     Lear
    -0.78
    washer
    -0.71
    ciating
    -0.70
    heid
    -0.69
    raint
    -0.68
    clamation
    -0.67
     Knowledge
    -0.66
     purity
    -0.63
    Appearance
    -0.62
    htaking
    -0.62
    POSITIVE LOGITS
     relocate
    1.06
     revert
    1.01
     postpone
    0.99
     sooner
    0.99
     renegoti
    0.97
     reconsider
    0.97
     reintrodu
    0.95
     reopen
    0.93
     withdraw
    0.91
     postp
    0.91
    Activations Density 0.319%

    No Known Activations