© 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. 9255
    Prev
    Next
    INDEX
    Explanations

    references to a specific individual named Masa

    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
    515
    +0.11
    0.5%
    57
    +0.11
    0.5%
    686
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    57
    +0.11
    0.01
    515
    +0.11
    0.01
    134
    +0.09
    0.01
    Negative Logits
    sburgh
    -0.73
    ship
    -0.72
    ski
    -0.69
    ï¸ı
    -0.67
     Ames
    -0.64
    orer
    -0.63
     BMC
    -0.63
    unes
    -0.62
     Saskatchewan
    -0.60
     Nordic
    -0.60
    POSITIVE LOGITS
    querade
    1.29
    cul
    1.22
    quer
    1.09
    ques
    1.05
    sey
    1.02
    dar
    0.99
    que
    0.98
    qu
    0.96
    ala
    0.93
    ih
    0.89
    Activations Density 0.019%

    No Known Activations