© 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. 12848
    Prev
    Next
    INDEX
    Explanations

    numbers and special characters in a non-English script, likely Russian

    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
    284
    +0.13
    0.6%
    649
    +0.10
    0.5%
    216
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    284
    +0.13
    0.02
    45
    +0.10
    0.01
    103
    +0.10
    0.01
    Negative Logits
    kson
    -0.78
    kins
    -0.75
    llan
    -0.73
    undai
    -0.72
    uke
    -0.67
    OHN
    -0.67
    pload
    -0.66
    merce
    -0.66
    gewater
    -0.66
    ffee
    -0.65
    POSITIVE LOGITS
    и
    1.06
    Ñĭ
    1.03
    ÑĤ
    0.98
    оÐ
    0.97
    а
    0.95
    о
    0.93
    ¤
    0.83
    ãĥĥãĥī
    0.82
    н
    0.82
    д
    0.82
    Activations Density 0.014%

    No Known Activations