© 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. 16912
    Prev
    Next
    INDEX
    Explanations

    phrases related to negotiation and decision-making

    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
    80
    +0.10
    0.5%
    469
    +0.10
    0.5%
    608
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    665
    +0.10
    0.07
    750
    +0.10
    0.05
    566
    +0.10
    0.06
    Negative Logits
    inas
    -0.69
    bis
    -0.68
    arta
    -0.66
    hene
    -0.58
     Dates
    -0.57
    icipated
    -0.56
    ogeneous
    -0.56
     TOD
    -0.56
    obyl
    -0.56
    aka
    -0.55
    POSITIVE LOGITS
     nonetheless
    1.07
     anyway
    0.93
     reluctantly
    0.92
     instinctively
    0.92
     anyways
    0.89
     accordingly
    0.88
    etheless
    0.80
     aback
    0.77
     gladly
    0.77
     tremend
    0.76
    Activations Density 0.520%

    No Known Activations