© 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. 901
    Prev
    Next
    INDEX
    Explanations

    terms related to technology and media ethics

    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
    314
    +0.11
    0.5%
    481
    +0.09
    0.4%
    127
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    317
    +0.11
    0.09
    621
    +0.09
    0.09
    740
    +0.09
    0.08
    Negative Logits
     shroud
    -0.81
     Belg
    -0.78
    wagen
    -0.77
     Manit
    -0.77
    creen
    -0.76
     Mous
    -0.75
     Reprodu
    -0.73
     guiActiveUnfocused
    -0.71
     museum
    -0.69
     division
    -0.67
    POSITIVE LOGITS
    ľ
    1.66
    ª
    1.37
    ł
    1.33
    Ķ
    1.29
    IJ
    1.28
    ¡
    1.23
    ¦
    1.18
    ij
    1.16
    ¤
    1.15
    ĺ
    1.15
    Activations Density 0.153%

    No Known Activations