© 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 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. 17525
    Prev
    Next
    INDEX
    Explanations

    phrases related to being forced or pressured to do 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_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
    665
    +0.11
    0.5%
    470
    +0.11
    0.5%
    105
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    470
    +0.11
    0.02
    665
    +0.11
    0.03
    585
    +0.10
    0.02
    Negative Logits
    vironment
    -0.78
    ership
    -0.72
     Offense
    -0.66
    amara
    -0.65
    izoph
    -0.65
    aming
    -0.63
     Enh
    -0.62
    itect
    -0.62
    aghd
    -0.61
    rou
    -0.61
    POSITIVE LOGITS
    otom
    0.88
     aback
    0.81
    untarily
    0.80
     shut
    0.78
     overtime
    0.77
     into
    0.74
     closure
    0.74
     resorted
    0.73
     compelled
    0.69
     dragged
    0.67
    Activations Density 0.031%

    No Known Activations