© 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. 9858
    Prev
    Next
    INDEX
    Explanations

    the word "Mall" or variations thereof

    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
    526
    +0.16
    0.7%
    288
    +0.13
    0.6%
    481
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.16
    0.03
    547
    +0.13
    0.02
    526
    +0.12
    0.00
    Negative Logits
    dit
    -0.69
    DIT
    -0.64
    IBLE
    -0.63
    SHIP
    -0.62
    vous
    -0.60
    lies
    -0.59
    ð
    -0.58
     Nich
    -0.58
     hottest
    -0.58
    QUI
    -0.57
    POSITIVE LOGITS
    adium
    0.96
    esi
    0.96
    ocks
    0.90
    ageddon
    0.87
    ieu
    0.87
    itect
    0.86
    iser
    0.84
    imar
    0.84
    igl
    0.84
     kios
    0.83
    Activations Density 0.079%

    No Known Activations