© 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. 23030
    Prev
    Next
    INDEX
    Explanations

    the word "for" in various contexts, highlighting its frequent use in sentences

    oai_token-act-pair · gpt-4o-miniTriggered 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
    24
    +0.10
    0.4%
    460
    +0.09
    0.4%
    335
    +0.08
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    24
    +0.10
    0.03
    362
    +0.09
    0.03
    335
    +0.08
    0.03
    Negative Logits
    ohn
    -0.72
    querque
    -0.68
    roach
    -0.66
    Ver
    -0.64
    itect
    -0.63
    quist
    -0.63
    panic
    -0.63
    alky
    -0.61
    gae
    -0.61
    geist
    -0.60
    POSITIVE LOGITS
     supremacy
    1.19
     scraps
    1.13
     dominance
    0.90
    gotten
    0.88
     survival
    0.88
     superiority
    0.83
     prizes
    0.80
    ties
    0.78
     scarce
    0.78
     custody
    0.77
    Activations Density 0.088%

    No Known Activations