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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsBlogSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 3072
    5. 8-RES_FS3072-JB
    6. 1267
    Prev
    Next
    INDEX
    Explanations

    political and legal terms

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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
    +0.28
    1.5%
    718
    +0.09
    0.5%
    702
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    87
    +0.28
    -0.10
    680
    +0.09
    0.10
    756
    +0.08
    -0.11
    Negative Logits
     Tycoon
    -0.61
     raping
    -0.56
     veget
    -0.55
    enery
    -0.55
     virginity
    -0.54
     predators
    -0.52
    animate
    -0.52
    ampires
    -0.52
     physically
    -0.49
     acceler
    -0.49
    POSITIVE LOGITS
     publication
    0.67
     quoting
    0.67
    ascript
    0.65
     citation
    0.65
     reader
    0.62
     Journals
    0.62
     explanatory
    0.61
     Editorial
    0.61
     commenting
    0.61
     journalistic
    0.60
    Activations Density 9.043%

    No Known Activations