© 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. 13569
    Prev
    Next
    INDEX
    Explanations

    text related to coding structures and functions

    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
    +0.42
    2.1%
    447
    +0.09
    0.4%
    672
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    480
    +0.42
    -0.00
    266
    +0.09
    0.10
    686
    +0.09
    0.04
    Negative Logits
     meanwhile
    -0.65
    ?]
    -0.64
    Again
    -0.64
    '?
    -0.64
     moreover
    -0.59
    )?
    -0.59
    Secondly
    -0.58
    ':
    -0.58
    ?'
    -0.58
     meantime
    -0.56
    POSITIVE LOGITS
     paranormal
    0.68
     environmentally
    0.66
     futuristic
    0.65
     iconic
    0.62
    imum
    0.61
     supernatural
    0.59
     prehistoric
    0.58
     customizable
    0.58
    uca
    0.58
     fictional
    0.57
    Activations Density 1.384%

    No Known Activations