© 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 12288
    9. 8-RES_FS12288-JB
    11. 9799
    Prev
    Next
    INDEX
    Explanations

    comparisons, especially using the word "than"

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs12288-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_12288
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    12,288
    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
    459
    +0.10
    0.5%
    393
    +0.09
    0.4%
    641
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    539
    +0.10
    0.04
    692
    +0.09
    0.04
    396
    +0.09
    0.04
    Negative Logits
    ALE
    -0.73
    EMENT
    -0.69
    uto
    -0.68
    urther
    -0.65
    raine
    -0.63
    obal
    -0.62
    erman
    -0.61
    ISSION
    -0.60
    eri
    -0.60
    rio
    -0.60
    POSITIVE LOGITS
     usual
    1.19
     anything
    1.00
     ever
    0.90
     anticipated
    0.88
     expected
    0.87
     anybody
    0.83
    average
    0.83
     average
    0.83
    usual
    0.82
     any
    0.82
    Activations Density 0.103%

    No Known Activations