© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 23855
    Prev
    Next
    INDEX
    Explanations

    references to records, milestones, or notable achievements

    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
    397
    +0.10
    0.5%
    252
    +0.09
    0.4%
    330
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    636
    +0.10
    0.02
    397
    +0.09
    0.02
    37
    +0.09
    0.02
    Negative Logits
    enegger
    -0.71
    enza
    -0.65
    uman
    -0.65
    eral
    -0.61
    lees
    -0.57
    ogyn
    -0.56
    unte
    -0.56
    ugg
    -0.56
    isner
    -0.55
    rieved
    -0.55
    POSITIVE LOGITS
     precedent
    1.03
     precedence
    1.03
     abl
    0.97
     sights
    0.91
    tle
    0.87
     groundwork
    0.84
    flame
    0.81
     preced
    0.81
     tone
    0.77
     benchmarks
    0.76
    Activations Density 0.087%

    No Known Activations