© 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. 9878
    Prev
    Next
    INDEX
    Explanations

    words related to politics and opinions regarding political figures

    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
    102
    +0.22
    1.1%
    55
    +0.18
    0.8%
    566
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    102
    +0.22
    0.15
    55
    +0.18
    0.14
    566
    +0.12
    0.10
    Negative Logits
     Moines
    -0.73
    umbnails
    -0.73
    ossom
    -0.73
    udeau
    -0.71
    onnaissance
    -0.71
    urry
    -0.70
    illon
    -0.68
    inav
    -0.68
     Clarks
    -0.66
     Letters
    -0.65
    POSITIVE LOGITS
    nt
    1.33
     indeed
    1.15
     destined
    1.08
     going
    1.05
     gonna
    1.03
     somehow
    1.01
     capable
    1.01
     incapable
    1.00
     not
    0.99
     worth
    0.96
    Activations Density 0.495%

    No Known Activations