© 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. 22712
    Prev
    Next
    INDEX
    Explanations

    elements related to optimization and enhancement in products or experiences

    oai_token-act-pair · gpt-4o-miniTriggered 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
    194
    +0.11
    0.5%
    647
    +0.10
    0.5%
    465
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    194
    +0.11
    0.04
    647
    +0.10
    0.04
    36
    +0.10
    0.05
    Negative Logits
     polls
    -0.80
     Journals
    -0.78
     jails
    -0.77
    arians
    -0.74
    writers
    -0.72
     tweets
    -0.72
     books
    -0.72
     surveys
    -0.71
     Methodist
    -0.70
     journals
    -0.70
    POSITIVE LOGITS
     versatility
    1.29
    imum
    1.11
     stiffness
    1.10
     airflow
    1.10
     aesthetic
    1.08
     smooth
    1.08
     tactile
    1.07
     optimum
    1.06
     flexibility
    1.03
     thickness
    1.02
    Activations Density 0.345%

    No Known Activations