© 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. 5238
    Prev
    Next
    INDEX
    Explanations

    information related to online activities and shopping

    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
    447
    +0.18
    0.8%
    525
    +0.11
    0.5%
    736
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    245
    +0.18
    0.03
    554
    +0.11
    0.01
    388
    +0.10
    0.03
    Negative Logits
    ppe
    -0.67
    itar
    -0.66
    IENT
    -0.63
    ariat
    -0.57
    chest
    -0.56
    itarian
    -0.56
     Nost
    -0.55
    yg
    -0.55
     Meter
    -0.53
    otta
    -0.52
    POSITIVE LOGITS
    without
    0.90
    aneously
    0.84
    with
    0.79
    within
    0.76
    eatures
    0.76
     without
    0.75
     cheaply
    0.75
     unnoticed
    0.74
    etheless
    0.74
    instead
    0.74
    Activations Density 0.126%

    No Known Activations