© 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 49152
    9. 8-RES_FS49152-JB
    11. 37171
    Prev
    Next
    INDEX
    Explanations

    references to nuclear events or conditions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    373
    +0.19
    0.9%
    702
    +0.12
    0.5%
    160
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    702
    +0.19
    0.03
    112
    +0.12
    0.02
    160
    +0.12
    0.02
    Negative Logits
    wagen
    -0.80
     lobb
    -0.67
     Elon
    -0.65
    nesday
    -0.63
    krit
    -0.63
     giveaways
    -0.62
     reinvest
    -0.62
    ullivan
    -0.61
     Fug
    -0.61
     UNCLASSIFIED
    -0.61
    POSITIVE LOGITS
     edit
    1.43
    src
    1.05
    ?]
    1.01
    actionDate
    1.00
    edit
    0.97
     Edit
    0.96
     citation
    0.93
    ËĪ
    0.93
    verified
    0.90
    note
    0.89
    Activations Density 0.011%

    No Known Activations