© 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. 19376
    Prev
    Next
    INDEX
    Explanations

    references to the musician 'Rock'

    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
    649
    +0.10
    0.5%
    325
    +0.10
    0.5%
    622
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    325
    +0.10
    0.02
    42
    +0.10
    0.02
    525
    +0.10
    0.02
    Negative Logits
    URES
    -0.79
    urers
    -0.65
    URE
    -0.65
    BILITIES
    -0.64
    unctions
    -0.61
    ples
    -0.60
     Chandra
    -0.60
     Breach
    -0.59
     practicable
    -0.59
    verages
    -0.59
    POSITIVE LOGITS
    castle
    1.01
    star
    1.01
    stars
    0.97
    ledge
    0.97
    ford
    0.93
    stead
    0.92
    ete
    0.90
    estone
    0.89
    cliffe
    0.89
    ers
    0.89
    Activations Density 0.017%

    No Known Activations