© 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. 815
    Prev
    Next
    INDEX
    Explanations

    words related to tradition or traditional methods

    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
    339
    +0.15
    0.7%
    288
    +0.13
    0.6%
    284
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    339
    +0.15
    0.03
    284
    +0.13
    0.03
    622
    +0.12
    0.02
    Negative Logits
     Wanted
    -0.76
     Imran
    -0.66
    upon
    -0.65
    hung
    -0.64
    atoon
    -0.64
     Aware
    -0.64
    mentation
    -0.64
     Allaah
    -0.61
    aughter
    -0.59
    robe
    -0.59
    POSITIVE LOGITS
     wisdom
    0.95
    arily
    0.95
    ised
    0.92
    ists
    0.92
    izations
    0.90
    ized
    0.88
    ties
    0.84
     notions
    0.81
    ization
    0.80
    isation
    0.78
    Activations Density 0.065%

    No Known Activations