© 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 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. 18105
    Prev
    Next
    INDEX
    Explanations

    cities

    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%
    607
    +0.12
    0.5%
    311
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    371
    +0.12
    0.04
    607
    +0.12
    0.03
    62
    +0.10
    0.03
    Negative Logits
    aceutical
    -0.68
    FTWARE
    -0.63
     Carmen
    -0.60
    Reviewer
    -0.56
     Giul
    -0.54
     kindly
    -0.53
     drums
    -0.53
     sul
    -0.53
     Marian
    -0.53
    Winged
    -0.53
    POSITIVE LOGITS
     least
    1.60
    onement
    1.31
    roph
    1.15
    yp
    1.02
    rial
    1.00
    hens
    0.97
    mosp
    0.92
    rophic
    0.90
    las
    0.90
    abase
    0.90
    Activations Density 0.075%

    No Known Activations