© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 7102
    Prev
    Next
    INDEX
    Explanations

    phrases related to language, accountability, and critical examination of societal and cultural dynamics

    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
    480
    +0.37
    1.9%
    447
    +0.27
    1.4%
    255
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    480
    +0.37
    -0.01
    665
    +0.27
    0.08
    567
    +0.11
    0.04
    Negative Logits
    itled
    -0.60
    Legend
    -0.59
    amon
    -0.56
    Lot
    -0.54
    sequently
    -0.52
    holiday
    -0.52
    aird
    -0.52
    iere
    -0.52
     Birthday
    -0.51
    oche
    -0.51
    POSITIVE LOGITS
     oneself
    1.17
     yourself
    0.90
     ourselves
    0.86
     objectively
    0.81
     Yourself
    0.78
     existing
    0.77
     passively
    0.75
     finite
    0.72
     yourselves
    0.69
     hypotheses
    0.66
    Activations Density 1.109%

    No Known Activations