© 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. 6254
    Prev
    Next
    INDEX
    Explanations

    phrase structures related to technical instructions or processes

    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
    87
    +0.32
    1.8%
    668
    +0.11
    0.6%
    756
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    566
    +0.32
    0.07
    87
    +0.11
    -0.02
    756
    +0.10
    -0.02
    Negative Logits
    )"
    -0.56
     Pigs
    -0.54
     Kills
    -0.52
    bury
    -0.50
    )."
    -0.50
     tho
    -0.49
    )]
    -0.48
    hate
    -0.46
    tsky
    -0.46
     Hate
    -0.45
    POSITIVE LOGITS
    pmwiki
    0.68
     conspicuous
    0.66
     unmist
    0.60
     conspic
    0.59
     indistinguishable
    0.59
     decidedly
    0.58
     noticeably
    0.57
     understandably
    0.56
     remarkably
    0.53
     unavoid
    0.52
    Activations Density 0.774%

    No Known Activations