© 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 1536
    9. 8-RES_FS1536-JB
    11. 18
    Prev
    Next
    INDEX
    Explanations

    statements related to advocating or promoting violence or controversial opinions

    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
    319
    +0.11
    0.5%
    348
    +0.11
    0.5%
    167
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.11
    0.09
    348
    +0.11
    0.09
    163
    +0.10
    0.07
    Negative Logits
     constitu
    -0.62
     Io
    -0.62
     Nun
    -0.61
     Bris
    -0.59
     Hera
    -0.59
    terday
    -0.58
     Bulg
    -0.57
    ãĤ¡
    -0.56
     Remem
    -0.56
     Isles
    -0.56
    POSITIVE LOGITS
    etermin
    1.37
    ividual
    1.32
    etermined
    1.27
    etermination
    1.23
    iverse
    1.21
    ownt
    1.21
    ynam
    1.20
    olph
    1.18
    imensional
    1.18
    rones
    1.17
    Activations Density 0.940%

    No Known Activations