© 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. 6596
    Prev
    Next
    INDEX
    Explanations

    terms related to medical procedures, equipment, and monitoring systems

    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
    481
    +0.21
    1.0%
    447
    +0.11
    0.5%
    34
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    388
    +0.21
    0.04
    34
    +0.11
    0.04
    620
    +0.10
    0.02
    Negative Logits
     Verse
    -0.76
     Estate
    -0.76
    erenn
    -0.74
    é»Ĵ
    -0.73
     Ruins
    -0.72
     Taxes
    -0.72
     Rebellion
    -0.70
     Territories
    -0.70
     Ethnic
    -0.69
    fest
    -0.69
    POSITIVE LOGITS
     sensors
    1.27
     detectors
    1.25
     scanners
    1.22
     detecting
    1.21
     malfunction
    1.19
     operated
    1.19
     monitors
    1.13
     detector
    1.12
     transmit
    1.11
     detects
    1.11
    Activations Density 0.710%

    No Known Activations