© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 1671
    Prev
    Next
    INDEX
    Explanations

    mentions of political or controversial issues with a negative connotation, especially related to specific organizations or individuals

    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
    369
    +0.13
    0.6%
    520
    +0.12
    0.6%
    354
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    373
    +0.13
    -0.02
    284
    +0.12
    0.04
    232
    +0.11
    0.04
    Negative Logits
     commissions
    -0.63
     trump
    -0.62
     preference
    -0.58
     opinion
    -0.58
    yip
    -0.58
    mot
    -0.57
    achus
    -0.57
     tart
    -0.56
    arching
    -0.56
     spor
    -0.56
    POSITIVE LOGITS
    «ĺ
    0.81
    ILE
    0.76
    orate
    0.75
    orius
    0.74
    ij士
    0.74
    Ĩ
    0.73
    ª
    0.71
    Ľ
    0.70
    CRIP
    0.69
    assi
    0.68
    Activations Density 0.252%

    No Known Activations