© 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. 7391
    Prev
    Next
    INDEX
    Explanations

    terms related to legal cases or criminal activities

    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
    481
    +0.17
    0.8%
    87
    +0.14
    0.6%
    679
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    452
    +0.17
    0.05
    165
    +0.14
    0.03
    679
    +0.10
    0.08
    Negative Logits
     ç¥ŀ
    -1.17
    Brexit
    -0.81
     masc
    -0.78
     haha
    -0.77
    Ïī
    -0.76
    iour
    -0.75
    âĻ
    -0.75
    pop
    -0.74
    Tomorrow
    -0.74
    ãĥİ
    -0.74
    POSITIVE LOGITS
     affidav
    1.32
     redacted
    1.32
     documents
    1.20
    Documents
    1.20
     disclosures
    1.18
     alleged
    1.15
     Documents
    1.14
     memos
    1.12
     subpoen
    1.12
     alleges
    1.11
    Activations Density 0.930%

    No Known Activations