© 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. 12487
    Prev
    Next
    INDEX
    Explanations

    mentions of legal proceedings and events related to business and technology

    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
    720
    +0.14
    0.7%
    458
    +0.11
    0.5%
    233
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    720
    +0.14
    0.07
    675
    +0.11
    0.06
    327
    +0.10
    0.06
    Negative Logits
    asser
    -0.80
    alus
    -0.80
    yz
    -0.77
    æ
    -0.75
    ijah
    -0.75
    Socket
    -0.73
    igo
    -0.72
    ois
    -0.72
    amar
    -0.71
    iera
    -0.71
    POSITIVE LOGITS
    assetsadobe
    0.69
    ï¸ı
    0.66
     antit
    0.66
     tolerate
    0.64
     deserve
    0.64
     preferring
    0.64
     wrists
    0.64
     refrain
    0.62
     restrooms
    0.62
     hardest
    0.62
    Activations Density 0.201%

    No Known Activations