© 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 12288
    9. 8-RES_FS12288-JB
    11. 1232
    Prev
    Next
    INDEX
    Explanations

    words related to loud sounds or noise

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    380
    +0.14
    0.6%
    96
    +0.10
    0.4%
    553
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    380
    +0.14
    0.02
    326
    +0.10
    0.02
    201
    +0.09
    0.02
    Negative Logits
    folios
    -0.81
     Reincarn
    -0.79
    folio
    -0.78
    utics
    -0.74
     Redemption
    -0.70
    clipse
    -0.69
     Journals
    -0.69
    photos
    -0.69
    hews
    -0.68
    akeru
    -0.68
    POSITIVE LOGITS
    mouth
    1.34
    eners
    1.14
     louder
    1.11
    ened
    1.04
    ness
    1.01
    ening
    1.00
     voices
    0.99
     noises
    0.99
     loud
    0.99
    ener
    0.97
    Activations Density 0.078%

    No Known Activations