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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 19793
    Prev
    Next
    INDEX
    Explanations

    proper nouns, specifically names of people, places, and organizations

    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
    659
    +0.21
    1.0%
    314
    +0.12
    0.6%
    197
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.21
    0.07
    197
    +0.12
    0.04
    67
    +0.12
    0.04
    Negative Logits
     Marketable
    -0.74
    terday
    -0.73
    ãĥ¼ãĥĨãĤ£
    -0.68
     wip
    -0.66
     courtesy
    -0.62
     lett
    -0.61
     ascending
    -0.61
     heads
    -0.61
     letting
    -0.60
     acknow
    -0.60
    POSITIVE LOGITS
    EEP
    1.18
    essel
    1.15
    ratom
    1.15
    udos
    1.10
    avan
    1.10
    aspers
    1.09
    ittens
    1.08
    ettle
    1.08
    ISS
    1.08
    ernels
    1.05
    Activations Density 0.034%

    No Known Activations