© 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. 12638
    Prev
    Next
    INDEX
    Explanations

    information related to crime or investigations

    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
    685
    +0.12
    0.5%
    763
    +0.10
    0.5%
    593
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    603
    +0.12
    0.04
    714
    +0.10
    0.03
    756
    +0.10
    -0.01
    Negative Logits
    DragonMagazine
    -0.86
    Picture
    -0.80
    mie
    -0.78
    quote
    -0.75
    DK
    -0.74
    assi
    -0.73
    aurus
    -0.72
    ahime
    -0.71
    Beast
    -0.70
    mire
    -0.68
    POSITIVE LOGITS
     themselves
    0.95
     individually
    0.86
     subsidiaries
    0.81
     clustered
    0.80
     residences
    0.79
     handguns
    0.77
     residential
    0.74
     households
    0.73
     locations
    0.73
     homes
    0.73
    Activations Density 0.403%

    No Known Activations