© 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. 2884
    Prev
    Next
    INDEX
    Explanations

    references to thoughts or concerns going through a person's mind

    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
    +0.45
    2.4%
    266
    +0.11
    0.6%
    172
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    480
    +0.45
    0.01
    266
    +0.11
    0.10
    172
    +0.10
    0.07
    Negative Logits
     (>
    -0.67
    hovah
    -0.66
    anwhile
    -0.66
     (<
    -0.65
    Therefore
    -0.64
    artment
    -0.62
    £ı
    -0.58
    cember
    -0.57
     Palest
    -0.56
     [*
    -0.56
    POSITIVE LOGITS
     unforgettable
    0.81
     fans
    0.70
    Fans
    0.68
     intrigue
    0.67
     sights
    0.67
     thrilled
    0.65
     eyebrows
    0.65
     surprises
    0.63
     accol
    0.62
     nightmares
    0.62
    Activations Density 1.224%

    No Known Activations