© 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. 11951
    Prev
    Next
    INDEX
    Explanations

    words related to specific topics such as solar power, investigations, and role-playing games

    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
    314
    +0.12
    0.5%
    96
    +0.10
    0.5%
    574
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.12
    0.09
    575
    +0.10
    0.06
    641
    +0.10
    0.05
    Negative Logits
    selves
    -0.88
    nesses
    -0.83
    terness
    -0.69
    ness
    -0.67
    thus
    -0.66
    cies
    -0.65
     Subtle
    -0.63
    Dialogue
    -0.63
    ingen
    -0.61
    clipse
    -0.60
    POSITIVE LOGITS
    tech
    0.78
    less
    0.77
     guiActiveUn
    0.77
     locker
    0.74
     boarding
    0.73
    ãĥ¼ãĥĨãĤ£
    0.71
    -
    0.69
    film
    0.68
     oriented
    0.68
     desk
    0.67
    Activations Density 0.465%

    No Known Activations