© 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 24576
    5. 8-RES_FS24576-JB
    6. 20084
    Prev
    Next
    INDEX
    Explanations

    keywords related to crafting items (crochet), medical terms (epilepsy, impairment), and DIY beauty-related activities (hair dyeing, hair care)

    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
    678
    +0.10
    0.4%
    71
    +0.09
    0.4%
    575
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.10
    0.10
    575
    +0.09
    0.09
    325
    +0.09
    0.07
    Negative Logits
    ertain
    -0.79
    pmwiki
    -0.69
    igious
    -0.65
    cellaneous
    -0.63
    addons
    -0.62
    bilt
    -0.62
    Minecraft
    -0.60
    rawdownloadcloneembedreportprint
    -0.60
    ktop
    -0.60
    alys
    -0.58
    POSITIVE LOGITS
    ensis
    0.84
     (<
    0.69
     displ
    0.67
     sandwic
    0.66
    tein
    0.65
    rique
    0.65
     (âĪĴ
    0.64
     preceded
    0.64
     Mk
    0.64
     paired
    0.63
    Activations Density 0.557%

    No Known Activations