© 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 EvalsExportsinterp-engineNEWAPI 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. 16888
    Prev
    Next
    INDEX
    Explanations

    instances related to building maintenance and safety issues

    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
    373
    +0.14
    0.7%
    114
    +0.11
    0.6%
    321
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    408
    +0.14
    0.09
    756
    +0.11
    -0.04
    79
    +0.10
    0.06
    Negative Logits
     Writ
    -0.68
    ®,
    -0.64
     wonderfully
    -0.62
     PLUS
    -0.61
     âĪ
    -0.60
     marvelous
    -0.59
     Beginning
    -0.58
     beautifully
    -0.58
    illac
    -0.58
    htaking
    -0.57
    POSITIVE LOGITS
    angering
    0.79
     counselling
    0.79
    Redd
    0.74
     witnesses
    0.72
     precautions
    0.72
    Investigators
    0.72
     perpetrators
    0.71
     incrim
    0.70
     attackers
    0.69
     victims
    0.69
    Activations Density 1.471%

    No Known Activations