© 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. 20537
    Prev
    Next
    INDEX
    Explanations

    specific keywords or phrases related to technical concepts or data analysis

    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
    653
    +0.11
    0.5%
    288
    +0.11
    0.5%
    418
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.11
    0.07
    653
    +0.11
    0.04
    541
    +0.09
    0.04
    Negative Logits
    cale
    -1.08
    aurus
    -0.88
    pite
    -0.83
    ynthesis
    -0.83
    tics
    -0.82
    acus
    -0.81
    hemy
    -0.81
    omen
    -0.78
    itch
    -0.78
    IGHTS
    -0.75
    POSITIVE LOGITS
     group
    0.81
     Worlds
    0.81
     runner
    0.78
     edition
    0.78
     dog
    0.77
     pilot
    0.76
     units
    0.75
     squad
    0.75
     vessel
    0.75
     batch
    0.75
    Activations Density 0.281%

    No Known Activations