© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 1564
    Prev
    Next
    INDEX
    Explanations

    references to firearms and gun-related issues

    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
    447
    +0.12
    0.6%
    314
    +0.11
    0.5%
    496
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    371
    +0.12
    0.03
    232
    +0.11
    0.02
    597
    +0.10
    0.02
    Negative Logits
     quo
    -0.76
     hor
    -0.63
     ABE
    -0.62
     impat
    -0.62
    ces
    -0.62
     otherwise
    -0.61
     estab
    -0.60
     toget
    -0.59
     oppos
    -0.59
     excuses
    -0.58
    POSITIVE LOGITS
    quartered
    1.11
    quarters
    1.03
    shake
    0.99
    sworth
    0.96
    igree
    0.96
    lers
    0.88
    wig
    0.87
    held
    0.86
    enstein
    0.86
    ley
    0.86
    Activations Density 0.064%

    No Known Activations