© 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. 18092
    Prev
    Next
    INDEX
    Explanations

    phrases related to online sign-ups, information, and notifications

    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
    97
    +0.14
    0.7%
    194
    +0.13
    0.6%
    285
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    97
    +0.14
    0.03
    194
    +0.13
    0.03
    62
    +0.09
    0.03
    Negative Logits
     Beng
    -0.66
    Pakistan
    -0.64
     Said
    -0.61
    ylum
    -0.61
     Came
    -0.60
     Kurd
    -0.59
     Jihad
    -0.58
    wat
    -0.58
     Mosque
    -0.58
     mosque
    -0.57
    POSITIVE LOGITS
     alerts
    0.81
     notifications
    0.78
     personalized
    0.78
     unlimited
    0.75
    irtual
    0.75
     FREE
    0.74
     instantly
    0.73
    exclusive
    0.72
     yourself
    0.71
    ospons
    0.71
    Activations Density 0.119%

    No Known Activations