© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 24257
    Prev
    Next
    INDEX
    Explanations

    instances of the word "about" indicating approximation or vague descriptions

    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
    260
    +0.13
    0.6%
    370
    +0.10
    0.5%
    270
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    260
    +0.13
    0.05
    724
    +0.10
    0.04
    405
    +0.10
    0.04
    Negative Logits
     Peaks
    -0.70
    hesis
    -0.68
    Express
    -0.68
    chens
    -0.66
    BT
    -0.62
    glas
    -0.62
    Write
    -0.61
     Writ
    -0.58
    NEY
    -0.58
    ocrine
    -0.58
    POSITIVE LOGITS
     halfway
    0.98
    aleb
    0.82
     midway
    0.79
     200
    0.75
     noon
    0.74
    inyl
    0.72
     eighty
    0.72
     Noon
    0.72
    etheless
    0.72
     850
    0.71
    Activations Density 0.073%

    No Known Activations