© 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. 2754
    Prev
    Next
    INDEX
    Explanations

    instances of the word "love" followed by a prepositional phrase

    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
    314
    +0.12
    0.6%
    469
    +0.11
    0.5%
    123
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    123
    +0.12
    0.04
    469
    +0.11
    0.03
    322
    +0.10
    0.03
    Negative Logits
    casters
    -0.73
    caster
    -0.66
     WOM
    -0.58
    RN
    -0.58
     Swordsman
    -0.56
     purch
    -0.56
     manag
    -0.55
     OWN
    -0.54
     ×
    -0.53
    Published
    -0.53
    POSITIVE LOGITS
     favor
    0.99
    escap
    0.95
     vain
    0.91
     favour
    0.91
     accordance
    0.90
    ordinate
    0.88
     unison
    0.84
    aus
    0.83
    offensive
    0.81
     line
    0.81
    Activations Density 0.083%

    No Known Activations