Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 1536
    5. 8-RES_FS1536-JB
    6. 1255
    Prev
    Next
    INDEX
    Explanations

    people's names, especially those related to various political figures, and organizations like Justice Department and FBI

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs1536-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    1,536
    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
    223
    +0.13
    0.6%
    141
    +0.10
    0.5%
    515
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    232
    +0.13
    0.10
    659
    +0.10
    0.09
    141
    +0.10
    0.07
    Negative Logits
    BALL
    -0.69
    liga
    -0.68
    PDATE
    -0.67
     haste
    -0.61
    REP
    -0.59
    MENTS
    -0.59
    uated
    -0.59
    âĺħ
    -0.59
     upkeep
    -0.59
     favour
    -0.57
    POSITIVE LOGITS
    athan
    1.24
    ston
    1.05
     Doe
    1.03
    nie
    1.01
     Cena
    0.98
    stown
    0.94
    hood
    0.91
     Hancock
    0.89
    stone
    0.89
    sen
    0.87
    Activations Density 1.478%

    No Known Activations