© 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. 12111
    Prev
    Next
    INDEX
    Explanations

    places or objects within specific locations, such as rooms or closets, that may be related to crimes or investigations

    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
    +0.57
    3.1%
    447
    +0.17
    0.9%
    266
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.57
    0.12
    266
    +0.17
    0.09
    480
    +0.13
    0.01
    Negative Logits
    igl
    -0.59
    kef
    -0.58
    ppa
    -0.55
    meier
    -0.53
    fman
    -0.51
    itz
    -0.51
    jah
    -0.50
    xon
    -0.49
     Kiw
    -0.49
    arnaev
    -0.49
    POSITIVE LOGITS
     itself
    0.71
     consists
    0.61
     herself
    0.57
     yourself
    0.56
     minus
    0.54
     encompasses
    0.53
    pedia
    0.52
     ourselves
    0.52
     himself
    0.51
     behaves
    0.51
    Activations Density 0.935%

    No Known Activations