Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 24576
    5. 8-RES_FS24576-JB
    6. 18966
    Prev
    Next
    INDEX
    Explanations

    mentions of immigration, officials, and hiring related to Bernie Sanders' presidential campaign

    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
    87
    +0.22
    1.1%
    756
    +0.13
    0.6%
    374
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    87
    +0.22
    -0.03
    756
    +0.13
    -0.03
    696
    +0.10
    0.05
    Negative Logits
    notations
    -0.63
     ILCS
    -0.59
     GOODMAN
    -0.54
    RESULTS
    -0.54
     suscept
    -0.51
     Blizz
    -0.50
     Canaver
    -0.49
    ////////////////////////////////
    -0.48
     Cups
    -0.48
    rences
    -0.48
    POSITIVE LOGITS
    ,...
    0.93
     ..."
    0.93
     })
    0.78
     Replay
    0.75
    ;}
    0.69
     });
    0.68
    '';
    0.68
    sic
    0.68
     </
    0.64
    \'
    0.62
    Activations Density 1.133%

    No Known Activations