© 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. 13503
    Prev
    Next
    INDEX
    Explanations

    numbers indicating a proportion or ratio in a text context

    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
    71
    +0.11
    0.5%
    218
    +0.10
    0.5%
    742
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    344
    +0.11
    0.04
    742
    +0.10
    0.03
    596
    +0.10
    0.04
    Negative Logits
     Brief
    -0.65
     Dull
    -0.62
    Pic
    -0.60
     lil
    -0.59
    pport
    -0.59
     majesty
    -0.59
     limbs
    -0.59
     Puzz
    -0.57
    icut
    -0.56
     Lover
    -0.56
    POSITIVE LOGITS
     (>
    1.03
     of
    0.94
    UID
    0.81
     (~
    0.78
    bang
    0.74
    ois
    0.68
     OF
    0.65
    ÃĽ
    0.64
     âĶľ
    0.63
    metry
    0.63
    Activations Density 0.086%

    No Known Activations