© 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. 24224
    Prev
    Next
    INDEX
    Explanations

    conjunctions and expressions that indicate connectedness or continuation in text

    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
    705
    +0.12
    0.5%
    587
    +0.10
    0.5%
    578
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    720
    +0.12
    0.07
    546
    +0.10
    0.05
    517
    +0.10
    0.05
    Negative Logits
     corrid
    -0.65
    rices
    -0.61
     stip
    -0.61
    attribute
    -0.58
    chieve
    -0.58
     synd
    -0.58
     consistency
    -0.58
    amide
    -0.57
    gements
    -0.57
    asty
    -0.57
    POSITIVE LOGITS
     ready
    1.08
     Alive
    0.95
     awaiting
    0.94
     waiting
    0.93
     surrounded
    0.90
    ACTED
    0.90
     prow
    0.81
    occupied
    0.80
     poised
    0.79
    ready
    0.78
    Activations Density 0.400%

    No Known Activations