© 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. 257
    Prev
    Next
    INDEX
    Explanations

    instructions on what tools or ingredients are required for a task or recipe

    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
    52
    +0.09
    0.4%
    373
    +0.09
    0.4%
    323
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    609
    +0.09
    0.04
    52
    +0.09
    0.04
    8
    +0.09
    0.04
    Negative Logits
    Ń·
    -0.72
     Dism
    -0.68
     Dou
    -0.64
     Democr
    -0.64
    speak
    -0.61
    RL
    -0.60
    Dialog
    -0.59
    grim
    -0.58
     Skies
    -0.58
     Sounds
    -0.58
    POSITIVE LOGITS
    lessly
    1.26
    nces
    0.79
     patience
    0.78
     recourse
    0.75
     permission
    0.73
    n
    0.72
     to
    0.71
    esome
    0.69
     assistance
    0.68
     authorization
    0.68
    Activations Density 0.052%

    No Known Activations