© 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
    2. Joseph Bloom · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 12288
    5. 8-RES_FS12288-JB
    6. 9516
    Prev
    Next
    INDEX
    Explanations

    descriptions of characters and their roles in a fictional world

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    480
    +0.47
    2.4%
    266
    +0.11
    0.5%
    447
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    480
    +0.47
    0.02
    266
    +0.11
    0.10
    543
    +0.10
    0.06
    Negative Logits
     Zucker
    -0.77
    espie
    -0.75
    ï¸ı
    -0.74
    ilater
    -0.71
    estyle
    -0.70
    TPPStreamerBot
    -0.70
    Hamilton
    -0.70
    mercial
    -0.70
     Greenberg
    -0.68
    Case
    -0.67
    POSITIVE LOGITS
     sorce
    1.15
     mages
    1.12
     Norn
    1.11
     undead
    1.10
     sorcerer
    1.09
     mage
    1.07
     Inquisitor
    1.07
     Undead
    1.06
     treacher
    1.05
     Aether
    1.05
    Activations Density 1.573%

    No Known Activations