© 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 12288
    9. 8-RES_FS12288-JB
    11. 618
    Prev
    Next
    INDEX
    Explanations

    web URLs and file paths

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    216
    +0.10
    0.5%
    430
    +0.10
    0.5%
    195
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    746
    +0.10
    0.02
    412
    +0.10
    0.02
    160
    +0.10
    0.03
    Negative Logits
     Beir
    -0.80
    terday
    -0.76
    glers
    -0.71
    itated
    -0.64
    hift
    -0.64
    inished
    -0.63
    onement
    -0.62
    idious
    -0.61
    neys
    -0.61
     Grimes
    -0.60
    POSITIVE LOGITS
    ËĪ
    1.48
    usr
    1.16
    Applications
    0.89
    etc
    0.88
     Via
    0.88
    Film
    0.83
    home
    0.82
    Ħ¢
    0.80
    wcsstore
    0.80
    u
    0.77
    Activations Density 0.046%

    No Known Activations