© 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. 23878
    Prev
    Next
    INDEX
    Explanations

    terms related to change or transformation

    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
    665
    +0.12
    0.5%
    4
    +0.11
    0.5%
    288
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    665
    +0.12
    0.06
    271
    +0.11
    0.04
    468
    +0.10
    0.04
    Negative Logits
    esp
    -0.59
    of
    -0.58
    avery
    -0.58
    saf
    -0.55
     guarantee
    -0.55
     Straw
    -0.55
     suggestions
    -0.54
     Wynne
    -0.54
    croft
    -0.54
    arov
    -0.54
    POSITIVE LOGITS
     peacefully
    0.97
     oneself
    0.89
     anew
    0.81
     explos
    0.80
     herself
    0.78
     accordingly
    0.78
     into
    0.78
     themselves
    0.77
     voluntarily
    0.77
     himself
    0.76
    Activations Density 0.236%

    No Known Activations