© 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 24576
    9. 8-RES_FS24576-JB
    11. 19589
    Prev
    Next
    INDEX
    Explanations

    information related to police officers, investigations, and incidents involving law enforcement

    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
    738
    +0.11
    0.5%
    43
    +0.11
    0.5%
    627
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    742
    +0.11
    0.03
    516
    +0.11
    0.03
    627
    +0.11
    0.02
    Negative Logits
    rencies
    -0.97
     currencies
    -0.89
     cryptocurrencies
    -0.88
     seeds
    -0.84
     Seeds
    -0.82
     UNESCO
    -0.80
     Centauri
    -0.77
     prophe
    -0.77
     iceberg
    -0.77
     billionaires
    -0.76
    POSITIVE LOGITS
     patrol
    1.28
     patrolling
    1.27
     Patrol
    1.21
     Officer
    1.11
     patrols
    1.07
     SWAT
    1.07
     handc
    1.06
     Sergeant
    1.06
     LAPD
    1.05
     handcuffed
    1.05
    Activations Density 0.545%

    No Known Activations