© 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. 22974
    Prev
    Next
    INDEX
    Explanations

    activities and actions described in a narrative context

    oai_token-act-pair · gpt-4o-miniTriggered 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
    87
    +0.20
    1.0%
    373
    +0.12
    0.6%
    756
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    665
    +0.20
    0.06
    87
    +0.12
    -0.03
    748
    +0.10
    0.03
    Negative Logits
    ador
    -0.62
     nowadays
    -0.57
     Historically
    -0.56
    zbollah
    -0.55
    obyl
    -0.54
     historically
    -0.54
     disliked
    -0.52
     lately
    -0.51
     or
    -0.50
    than
    -0.49
    POSITIVE LOGITS
     accordingly
    0.71
    FIR
    0.64
     proceeded
    0.62
     proble
    0.60
     [|
    0.60
     prest
    0.57
    ãĤ©
    0.56
    CLASSIFIED
    0.56
     vo
    0.56
     mosqu
    0.54
    Activations Density 0.754%

    No Known Activations