© 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 49152
    9. 8-RES_FS49152-JB
    11. 45809
    Prev
    Next
    INDEX
    Explanations

    references to various types of sensors

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    575
    +0.16
    0.8%
    288
    +0.16
    0.8%
    578
    +0.15
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    575
    +0.16
    0.02
    578
    +0.16
    0.01
    332
    +0.15
    0.01
    Negative Logits
    swick
    -0.86
    ership
    -0.82
    heit
    -0.76
    tions
    -0.75
     Masquerade
    -0.69
     Bald
    -0.68
    ingo
    -0.68
     Klux
    -0.67
    hammer
    -0.66
     Alban
    -0.65
    POSITIVE LOGITS
     sensors
    1.03
     sensor
    0.96
     detects
    0.91
     calibration
    0.86
     subsystem
    0.85
    Sensor
    0.83
     microphones
    0.81
    ionics
    0.79
     scanner
    0.78
    igr
    0.78
    Activations Density 0.006%

    No Known Activations