© 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. 2391
    Prev
    Next
    INDEX
    Explanations

    discussions or statements related to people's opinions and perspectives

    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
    447
    +0.14
    0.6%
    19
    +0.11
    0.5%
    645
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    455
    +0.14
    0.05
    349
    +0.11
    0.03
    722
    +0.10
    0.03
    Negative Logits
    =~
    -0.74
    -+
    -0.74
    baugh
    -0.72
    psc
    -0.70
    ĸļ
    -0.70
     Anyone
    -0.70
    ãĤ´
    -0.68
    --------------------------------------------------------
    -0.67
    ————————
    -0.67
    Anyone
    -0.65
    POSITIVE LOGITS
     downright
    0.87
     outright
    0.80
     prefer
    0.74
     even
    0.74
     specialize
    0.72
     worse
    0.69
     incorrectly
    0.69
     overlap
    0.69
     misinterpret
    0.65
     affiliate
    0.64
    Activations Density 0.265%

    No Known Activations