© 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. 18131
    Prev
    Next
    INDEX
    Explanations

    items of clothing and personal belongings associated with specific characters or TV shows

    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
    454
    +0.12
    0.6%
    760
    +0.12
    0.5%
    232
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    732
    +0.12
    0.03
    760
    +0.12
    0.03
    284
    +0.09
    0.03
    Negative Logits
    avorite
    -0.90
    nels
    -0.86
    hower
    -0.76
    keyes
    -0.74
     tow
    -0.68
    washer
    -0.68
     dipping
    -0.68
    roo
    -0.68
    cone
    -0.68
    usercontent
    -0.67
    POSITIVE LOGITS
     Aging
    1.19
     Madness
    1.10
     Politics
    1.09
     Strategies
    1.07
     Behavior
    1.06
     Terrorism
    1.05
     Tactics
    1.02
     Religion
    1.01
     Culture
    1.01
     Integrity
    1.01
    Activations Density 0.097%

    No Known Activations