© 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. 18392
    Prev
    Next
    INDEX
    Explanations

    descriptions of potential risks and consequences related to various situations or decisions

    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
    756
    +0.14
    0.7%
    509
    +0.10
    0.5%
    636
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    756
    +0.14
    -0.01
    636
    +0.10
    0.04
    570
    +0.09
    0.01
    Negative Logits
     lately
    -0.72
     costumes
    -0.69
     costume
    -0.63
     accents
    -0.63
     attire
    -0.62
     etiquette
    -0.61
     humour
    -0.61
    iosity
    -0.58
     fame
    -0.56
     visualization
    -0.55
    POSITIVE LOGITS
     indefinitely
    0.79
     permanently
    0.76
    ivably
    0.74
     2021
    0.71
    ãĤ°
    0.70
     poorer
    0.68
     forever
    0.68
     moot
    0.67
     2020
    0.66
     2024
    0.66
    Activations Density 0.950%

    No Known Activations