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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 17579
    Prev
    Next
    INDEX
    Explanations

    phrases indicating evaluation, particularly emphasizing minimum standards or negative judgment

    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
    756
    +0.11
    0.5%
    540
    +0.10
    0.4%
    705
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    43
    +0.11
    0.01
    553
    +0.10
    0.01
    705
    +0.09
    0.01
    Negative Logits
    borg
    -0.93
    ses
    -0.82
    nel
    -0.79
    axter
    -0.76
    kefeller
    -0.75
    alist
    -0.74
    ilater
    -0.71
    ocide
    -0.69
    olor
    -0.69
    byter
    -0.68
    POSITIVE LOGITS
     implying
    0.76
     bit
    0.72
     considering
    0.68
     impression
    0.67
     Pradesh
    0.65
     provocation
    0.65
     Abe
    0.64
     inviting
    0.64
     suggesting
    0.63
     spoiler
    0.62
    Activations Density 0.017%

    No Known Activations