© 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. 24028
    Prev
    Next
    INDEX
    Explanations

    place names and their significance to various contexts

    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
    481
    +0.24
    1.1%
    572
    +0.12
    0.6%
    312
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.24
    0.08
    481
    +0.12
    0.07
    146
    +0.11
    0.05
    Negative Logits
    ĪĴ
    -0.76
     classmate
    -0.59
    CHAT
    -0.59
     Cosponsors
    -0.58
    rompt
    -0.57
    URL
    -0.56
     subcontract
    -0.55
    ļé
    -0.55
    affle
    -0.54
    actionDate
    -0.54
    POSITIVE LOGITS
     since
    0.83
     worldwide
    0.73
    .
    0.72
     ever
    0.72
     insofar
    0.68
     imaginable
    0.68
     globally
    0.67
    ankind
    0.65
    .(
    0.64
    }.
    0.64
    Activations Density 0.489%

    No Known Activations