© 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 24576
    9. 8-RES_FS24576-JB
    11. 9319
    Prev
    Next
    INDEX
    Explanations

    words related to political pressure and conflict

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    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
    96
    +0.10
    0.5%
    458
    +0.10
    0.4%
    685
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    471
    +0.10
    0.03
    557
    +0.10
    0.03
    587
    +0.09
    0.04
    Negative Logits
    SourceFile
    -0.93
    flies
    -0.78
    olf
    -0.77
    pac
    -0.71
    ONSORED
    -0.70
    Ü
    -0.68
     Klux
    -0.67
    ¯¯¯¯
    -0.66
    consider
    -0.65
    aird
    -0.65
    POSITIVE LOGITS
    erous
    1.60
     behalf
    1.53
     unsuspecting
    1.08
    etime
    0.88
    shore
    0.87
     innocent
    0.76
     campuses
    0.76
     unwitting
    0.75
     top
    0.74
    anism
    0.73
    Activations Density 0.137%

    No Known Activations