© 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 3072
    9. 8-RES_FS3072-JB
    11. 2997
    Prev
    Next
    INDEX
    Explanations

    phrases related to legal proceedings and law enforcement

    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
    314
    +0.12
    0.6%
    373
    +0.12
    0.5%
    618
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    100
    +0.12
    0.07
    618
    +0.12
    0.08
    69
    +0.11
    0.06
    Negative Logits
     giveaway
    -0.66
     latch
    -0.55
     fractures
    -0.53
     giveaways
    -0.53
     obscurity
    -0.52
     controversies
    -0.51
     ledger
    -0.51
     Bast
    -0.51
    hawks
    -0.50
     trouble
    -0.49
    POSITIVE LOGITS
     "[
    1.18
     "...
    1.15
     "(
    1.09
     "âĢ¦
    1.08
     '[
    1.07
     "{
    1.05
     ""
    1.03
     "'
    1.01
     "@
    1.01
     "\
    0.91
    Activations Density 0.786%

    No Known Activations