© 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. 5517
    Prev
    Next
    INDEX
    Explanations

    the word 'to' followed by a numerical value indicating an action

    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
    718
    +0.14
    0.7%
    665
    +0.14
    0.7%
    49
    +0.13
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    49
    +0.14
    0.04
    665
    +0.14
    0.04
    718
    +0.13
    0.04
    Negative Logits
    soDeliveryDate
    -0.83
    rising
    -0.74
    lies
    -0.74
    ©¶æ¥µ
    -0.69
    quickShipAvailable
    -0.68
     Kinnikuman
    -0.68
    ById
    -0.67
     Said
    -0.66
     Says
    -0.64
    ÃĥÃĤÃĥÃĤ
    -0.63
    POSITIVE LOGITS
     maximize
    1.12
     emulate
    1.12
     participate
    1.10
     pursue
    1.10
     partake
    1.08
     avoid
    1.07
     indulge
    1.05
     explore
    1.05
     keep
    1.05
     stay
    1.04
    Activations Density 0.072%

    No Known Activations