© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 15627
    Prev
    Next
    INDEX
    Explanations

    uncomfortable situations or potential negative consequences associated with certain actions or decisions

    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
    488
    +0.10
    0.4%
    195
    +0.09
    0.4%
    270
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    511
    +0.10
    0.02
    488
    +0.09
    0.02
    13
    +0.09
    0.02
    Negative Logits
    LINE
    -0.77
    eve
    -0.74
    essee
    -0.73
    geist
    -0.72
    speak
    -0.72
    ricks
    -0.70
    dies
    -0.70
    ele
    -0.68
    aters
    -0.68
     Cosponsors
    -0.67
    POSITIVE LOGITS
     losing
    0.97
     extinction
    0.93
     jeopard
    0.92
     exposing
    0.89
     endanger
    0.86
     risking
    0.85
    lessly
    0.84
     sacrificing
    0.83
     embarrassment
    0.80
     ruining
    0.79
    Activations Density 0.036%

    No Known Activations