© 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. 2883
    Prev
    Next
    INDEX
    Explanations

    mentions of legal actions, law enforcement, and controversial events

    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
    516
    +0.14
    0.7%
    127
    +0.12
    0.6%
    87
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    516
    +0.14
    0.08
    127
    +0.12
    0.05
    288
    +0.12
    0.11
    Negative Logits
     anecd
    -0.72
     skelet
    -0.71
    theless
    -0.70
     pse
    -0.64
    DragonMagazine
    -0.63
     UNCLASSIFIED
    -0.63
    âĹ¼
    -0.63
     tremend
    -0.59
     looph
    -0.58
    arger
    -0.57
    POSITIVE LOGITS
     Donald
    0.66
    î
    0.61
    ADVERTISEMENT
    0.61
    ".
    0.59
     âĢİ
    0.59
     frontman
    0.59
     Belfast
    0.58
     [âĢ¦]
    0.58
    AFP
    0.56
    '.
    0.54
    Activations Density 1.702%

    No Known Activations