© 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. 9084
    Prev
    Next
    INDEX
    Explanations

    references to official websites

    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
    251
    +0.12
    0.5%
    377
    +0.11
    0.5%
    115
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    232
    +0.12
    0.01
    26
    +0.11
    0.01
    343
    +0.10
    0.01
    Negative Logits
    words
    -0.77
    teen
    -0.67
    DonaldTrump
    -0.67
    ciples
    -0.61
    scene
    -0.58
     trespass
    -0.58
    PLIED
    -0.57
    qus
    -0.56
    Fine
    -0.56
     downside
    -0.56
    POSITIVE LOGITS
    ensen
    1.05
    sky
    1.04
    enson
    0.99
    asms
    0.98
    roup
    0.98
    ues
    0.94
    ersen
    0.94
    uin
    0.91
    hetti
    0.89
    nette
    0.88
    Activations Density 0.016%

    No Known Activations