© 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. 16914
    Prev
    Next
    INDEX
    Explanations

    email subscription-related phrases

    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
    149
    +0.12
    0.6%
    194
    +0.11
    0.5%
    715
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    194
    +0.12
    0.03
    59
    +0.11
    0.03
    62
    +0.10
    0.02
    Negative Logits
    inventoryQuantity
    -0.75
     killings
    -0.72
    tight
    -0.69
    Phill
    -0.68
    hatt
    -0.65
    viol
    -0.64
    uph
    -0.63
    ese
    -0.62
    udi
    -0.61
    Mos
    -0.59
    POSITIVE LOGITS
     refunds
    1.12
     refund
    0.94
     reimb
    0.90
     unlimited
    0.85
     rewarded
    0.83
    imentary
    0.81
     rewards
    0.79
     discounted
    0.78
     bonus
    0.78
     bonuses
    0.76
    Activations Density 0.085%

    No Known Activations