© 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. 18834
    Prev
    Next
    INDEX
    Explanations

    phrases related to criminal behavior 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_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
    471
    +0.12
    0.6%
    356
    +0.10
    0.5%
    392
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    471
    +0.12
    0.02
    392
    +0.10
    0.02
    732
    +0.10
    0.02
    Negative Logits
    DAY
    -0.74
     Madness
    -0.71
    sth
    -0.69
     Elves
    -0.67
     Boll
    -0.67
     Archdemon
    -0.66
     Hour
    -0.65
     livest
    -0.63
    )=(
    -0.61
     attachments
    -0.61
    POSITIVE LOGITS
    ensible
    1.27
    ibr
    1.23
    erence
    1.22
    acement
    1.18
    erent
    1.16
    ected
    1.14
    ocused
    1.10
    ection
    1.09
    amiliar
    1.07
    endants
    1.06
    Activations Density 0.008%

    No Known Activations