© 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. 1180
    Prev
    Next
    INDEX
    Explanations

    mentions of veto actions or threats

    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
    218
    +0.12
    0.5%
    116
    +0.10
    0.4%
    738
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.12
    0.03
    500
    +0.10
    0.02
    326
    +0.09
    0.02
    Negative Logits
    ndra
    -0.83
    aster
    -0.73
    arers
    -0.71
    adr
    -0.70
    geist
    -0.70
     Occupations
    -0.69
     MPG
    -0.68
    brate
    -0.68
    istics
    -0.67
    asma
    -0.67
    POSITIVE LOGITS
     veto
    0.99
     vetoed
    0.87
     amnesty
    0.78
    æĢ
    0.74
    cies
    0.74
    zai
    0.73
     decree
    0.65
    ingen
    0.65
     bluff
    0.65
    es
    0.63
    Activations Density 0.150%

    No Known Activations