© 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 1536
    9. 8-RES_FS1536-JB
    11. 1040
    Prev
    Next
    INDEX
    Explanations

    words related to authoritarianism and control tactics

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs1536-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    1,536
    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
    167
    +0.10
    0.5%
    140
    +0.09
    0.4%
    659
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.10
    0.12
    232
    +0.09
    0.12
    326
    +0.09
    0.11
    Negative Logits
     kHz
    -0.67
    kHz
    -0.65
    birth
    -0.61
    terday
    -0.61
    éĹĺ
    -0.60
    ignty
    -0.59
     Tsu
    -0.59
    side
    -0.59
     Divinity
    -0.58
    CLASSIFIED
    -0.58
    POSITIVE LOGITS
    glers
    1.61
    gers
    1.54
    ging
    1.42
    gy
    1.35
    roup
    1.31
    lia
    1.30
    lio
    1.27
    ged
    1.24
    allery
    1.23
    raphics
    1.21
    Activations Density 2.255%

    No Known Activations