© 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. 1448
    Prev
    Next
    INDEX
    Explanations

    text related to legal and political concepts and actions

    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
    36
    +0.19
    0.9%
    393
    +0.14
    0.7%
    578
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    36
    +0.19
    0.12
    636
    +0.14
    0.06
    250
    +0.11
    0.06
    Negative Logits
    enance
    -0.97
    umin
    -0.79
    ighed
    -0.78
    times
    -0.77
    words
    -0.76
    govtrack
    -0.72
     onwards
    -0.70
    always
    -0.69
     aloud
    -0.67
    autions
    -0.67
    POSITIVE LOGITS
     makeshift
    1.02
     new
    0.97
     cohesive
    0.94
     viable
    0.89
     believable
    0.88
     roadmap
    0.86
     reputation
    0.85
     rapport
    0.85
     rudimentary
    0.84
     unified
    0.82
    Activations Density 0.343%

    No Known Activations