© 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. 17231
    Prev
    Next
    INDEX
    Explanations

    Russian characters and symbols, possibly associated with a specific language or encoding format

    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
    288
    +0.15
    0.7%
    45
    +0.12
    0.5%
    442
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    284
    +0.15
    0.01
    45
    +0.12
    0.01
    52
    +0.10
    0.01
    Negative Logits
    uyomi
    -0.76
     livest
    -0.76
     Skydragon
    -0.71
     conduc
    -0.68
    matically
    -0.68
    points
    -0.66
    ngth
    -0.65
    ippi
    -0.65
    hner
    -0.65
     whales
    -0.64
    POSITIVE LOGITS
    Ñĭ
    1.08
    ÑĢ
    0.98
    س
    0.98
    ÙĪ
    0.97
    л
    0.97
    Ñĥ
    0.96
    ãĥĥãĤ¯
    0.95
    ãĥ¥
    0.92
    о
    0.91
    rique
    0.89
    Activations Density 0.008%

    No Known Activations