© 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. 4404
    Prev
    Next
    INDEX
    Explanations

    phrases related to people being involved or participating in activities or events

    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
    497
    +0.13
    0.6%
    91
    +0.10
    0.5%
    32
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    91
    +0.13
    0.03
    377
    +0.10
    0.03
    368
    +0.10
    0.02
    Negative Logits
    ahime
    -0.90
    OUGH
    -0.69
    rar
    -0.66
    ijn
    -0.66
    LOCK
    -0.66
    haar
    -0.65
    Quotes
    -0.64
    Quantity
    -0.64
    rid
    -0.62
    appropriate
    -0.61
    POSITIVE LOGITS
     knew
    0.92
     interviewed
    0.86
     knows
    0.84
     alike
    0.83
     reacted
    0.83
     gasped
    0.82
     owe
    0.77
     whom
    0.77
     who
    0.76
     contacted
    0.75
    Activations Density 0.130%

    No Known Activations