© 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. 6216
    Prev
    Next
    INDEX
    Explanations

    references to events or actions that happen at a later time

    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
    410
    +0.10
    0.4%
    281
    +0.09
    0.4%
    430
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    410
    +0.10
    0.03
    237
    +0.09
    0.02
    281
    +0.09
    0.02
    Negative Logits
    DonaldTrump
    -0.71
    /+
    -0.69
    mad
    -0.68
    PK
    -0.67
    JV
    -0.66
    lua
    -0.66
     Pwr
    -0.65
    ignment
    -0.65
    THING
    -0.64
    KING
    -0.63
    POSITIVE LOGITS
     regretted
    0.99
    ally
    0.98
     clarified
    0.86
    onymous
    0.84
     deleted
    0.80
    forth
    0.79
    ality
    0.79
     apologized
    0.78
     discovered
    0.78
     succumbed
    0.77
    Activations Density 0.034%

    No Known Activations