© 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. 18801
    Prev
    Next
    INDEX
    Explanations

    repetitive occurrences of the word "again" in the text

    oai_token-act-pair · gpt-3.5-turboTriggered 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
    724
    +0.12
    0.6%
    71
    +0.11
    0.5%
    0
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    724
    +0.12
    0.02
    490
    +0.11
    0.02
    360
    +0.10
    0.02
    Negative Logits
    ocene
    -0.76
    76561
    -0.74
    regate
    -0.69
    owsky
    -0.68
    adena
    -0.66
     Institution
    -0.62
     Treaty
    -0.61
    liest
    -0.60
    oming
    -0.60
    rection
    -0.60
    POSITIVE LOGITS
    forth
    0.79
    nces
    0.73
    RGB
    0.72
    ndra
    0.70
     illustrating
    0.67
    mable
    0.67
    wcs
    0.66
     conclud
    0.66
    nir
    0.64
     proves
    0.64
    Activations Density 0.028%

    No Known Activations