© 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
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 24576
    5. 8-RES_FS24576-JB
    6. 18429
    Prev
    Next
    INDEX
    Explanations

    the specific word "Por" in the text

    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
    59
    +0.13
    0.6%
    699
    +0.11
    0.5%
    361
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    699
    +0.13
    0.01
    50
    +0.11
    0.01
    702
    +0.10
    0.01
    Negative Logits
    enegger
    -0.77
     GOODMAN
    -0.72
    ITNESS
    -0.72
    DERR
    -0.70
    EngineDebug
    -0.69
    INESS
    -0.68
    schild
    -0.68
     sorts
    -0.64
     Radiation
    -0.63
    ARDS
    -0.61
    POSITIVE LOGITS
    oshenko
    1.16
    ters
    1.10
    cup
    1.04
    cel
    1.02
    folios
    1.01
    plet
    1.01
    quet
    1.00
    bably
    1.00
    por
    0.99
    leased
    0.98
    Activations Density 0.015%

    No Known Activations