© 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
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 3072
    5. 8-RES_FS3072-JB
    6. 86
    Prev
    Next
    INDEX
    Explanations

    commands or phrases related to cutting something, shaping behavior, or getting rid of something

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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
    214
    +0.11
    0.5%
    566
    +0.10
    0.4%
    35
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    214
    +0.11
    0.04
    35
    +0.10
    0.04
    500
    +0.09
    0.04
    Negative Logits
     Monitor
    -0.69
     Moder
    -0.67
    Purchase
    -0.65
     Dunham
    -0.63
     Organisation
    -0.63
     Refresh
    -0.62
    SPONSORED
    -0.61
    AMP
    -0.60
     Kamp
    -0.60
     Franklin
    -0.59
    POSITIVE LOGITS
    scenes
    1.33
    thro
    1.32
    scene
    1.31
    lasses
    1.22
    aneous
    1.21
    lass
    1.05
    backs
    0.97
    eness
    0.93
    away
    0.92
    tle
    0.92
    Activations Density 0.714%

    No Known Activations