© 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. 10772
    Prev
    Next
    INDEX
    Explanations

    mentions or references to a specific person named "Kasim."

    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
    526
    +0.14
    0.7%
    383
    +0.11
    0.5%
    190
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.14
    0.01
    343
    +0.11
    0.01
    476
    +0.09
    0.01
    Negative Logits
     Islanders
    -0.83
    ODUCT
    -0.74
     exerc
    -0.66
    ycle
    -0.66
    UCT
    -0.65
    ANCE
    -0.64
     poppy
    -0.63
    ����
    -0.59
    Nap
    -0.59
     socket
    -0.59
    POSITIVE LOGITS
    imir
    1.06
    daq
    1.03
    iewicz
    1.02
    ala
    1.00
    bah
    0.97
    rils
    0.97
    erman
    0.92
    uga
    0.90
    idy
    0.90
    uth
    0.89
    Activations Density 0.016%

    No Known Activations