© 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. 6989
    Prev
    Next
    INDEX
    Explanations

    mathematician or architect names related to historical building and construction descriptions

    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
    +0.12
    0.6%
    235
    +0.12
    0.5%
    104
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    235
    +0.12
    0.05
    575
    +0.12
    0.05
    104
    +0.12
    0.04
    Negative Logits
     Pradesh
    -0.89
     Territories
    -0.76
    ibaba
    -0.74
     quo
    -0.66
     Elections
    -0.65
    topic
    -0.65
    plings
    -0.65
    estate
    -0.64
    inki
    -0.63
    jiang
    -0.62
    POSITIVE LOGITS
     extraord
    1.26
     laureate
    1.16
     specializing
    1.01
     Richard
    0.98
     Robert
    0.94
     Stephen
    0.92
     who
    0.90
     Philip
    0.86
     Charles
    0.86
     Nicholas
    0.85
    Activations Density 0.171%

    No Known Activations