© 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. 6829
    Prev
    Next
    INDEX
    Explanations

    experiences or situations

    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
    447
    +0.11
    0.5%
    101
    +0.10
    0.5%
    496
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    479
    +0.11
    0.01
    577
    +0.10
    0.02
    380
    +0.10
    0.01
    Negative Logits
    hatt
    -0.82
    enta
    -0.77
    law
    -0.76
    ature
    -0.74
    loader
    -0.74
    corn
    -0.70
    ulum
    -0.70
    loaded
    -0.70
    quote
    -0.70
    pta
    -0.67
    POSITIVE LOGITS
     firsthand
    1.30
     difficulties
    0.92
     hardships
    0.90
     homelessness
    0.87
     bouts
    0.82
     setbacks
    0.82
     debilitating
    0.81
     visions
    0.80
     frustrations
    0.79
     hardship
    0.75
    Activations Density 0.037%

    No Known Activations