© 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. 17060
    Prev
    Next
    INDEX
    Explanations

    features and qualities related to electronic devices

    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
    480
    +0.20
    0.9%
    420
    +0.12
    0.6%
    189
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    189
    +0.20
    0.05
    647
    +0.12
    0.04
    255
    +0.12
    0.04
    Negative Logits
     tours
    -0.83
     Indies
    -0.81
     lifestyles
    -0.79
     climates
    -0.75
     Towns
    -0.75
     Parties
    -0.73
     Tours
    -0.72
     documentaries
    -0.72
     riots
    -0.70
     jails
    -0.70
    POSITIVE LOGITS
     protr
    0.94
    binding
    0.88
     binding
    0.85
     fingerprint
    0.84
     signature
    0.83
    hidden
    0.83
     amulet
    0.83
     glyph
    0.82
     "+
    0.81
     socket
    0.81
    Activations Density 0.717%

    No Known Activations