© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 4530
    Prev
    Next
    INDEX
    Explanations

    words related to rewards or bonuses

    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
    763
    +0.15
    0.7%
    647
    +0.10
    0.4%
    32
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    763
    +0.15
    0.02
    647
    +0.10
    0.01
    32
    +0.09
    0.02
    Negative Logits
    adr
    -0.83
    stem
    -0.72
    prototype
    -0.69
    dfx
    -0.68
    ichen
    -0.66
    ugu
    -0.66
    ruary
    -0.66
    athed
    -0.66
    ancing
    -0.65
    ondon
    -0.65
    POSITIVE LOGITS
    cules
    0.80
    ++++
    0.77
    ï¸
    0.75
    henko
    0.66
     Scotia
    0.65
     Plenty
    0.65
     Zombies
    0.65
    leep
    0.63
     surprises
    0.63
     Clancy
    0.62
    Activations Density 0.015%

    No Known Activations