© 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. 9800
    Prev
    Next
    INDEX
    Explanations

    written text that describes or cites content from other sources

    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
    220
    +0.10
    0.5%
    323
    +0.10
    0.5%
    679
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    498
    +0.10
    0.04
    679
    +0.10
    0.05
    39
    +0.10
    0.04
    Negative Logits
    vp
    -0.74
    advertising
    -0.71
    thood
    -0.68
    nels
    -0.67
    ndra
    -0.67
    Sync
    -0.65
    akuya
    -0.65
    soDeliveryDate
    -0.63
    ometers
    -0.62
    FTWARE
    -0.62
    POSITIVE LOGITS
     excerpts
    1.32
     ital
    1.13
     excerpt
    1.13
     reprinted
    1.13
     redacted
    1.10
     quotations
    1.10
     reprint
    1.09
     condensed
    1.08
     transcript
    1.07
     reproduced
    1.04
    Activations Density 0.339%

    No Known Activations