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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 14643
    Prev
    Next
    INDEX
    Explanations

    references to intelligence and military activities, including specific details and organizations involved

    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
    679
    +0.12
    0.6%
    146
    +0.12
    0.6%
    109
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    146
    +0.12
    0.05
    682
    +0.12
    0.03
    288
    +0.10
    0.04
    Negative Logits
    Brew
    -0.82
     unborn
    -0.76
    ð
    -0.76
    à¼
    -0.76
    Whe
    -0.76
    Interstitial
    -0.76
    natureconservancy
    -0.74
     Towns
    -0.71
     Rivers
    -0.71
    Baby
    -0.69
    POSITIVE LOGITS
     spy
    1.23
     spies
    1.21
     intelligence
    1.17
     spying
    1.14
     KGB
    1.14
     operatives
    1.10
     espionage
    1.08
    elligence
    1.08
     interrogation
    1.07
     Intelligence
    1.04
    Activations Density 0.145%

    No Known Activations