© 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. 10443
    Prev
    Next
    INDEX
    Explanations

    proper nouns or names

    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
    668
    +0.11
    0.5%
    616
    +0.10
    0.5%
    364
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.11
    0.05
    284
    +0.10
    0.04
    28
    +0.10
    0.04
    Negative Logits
    ufact
    -0.86
    angelo
    -0.82
    sylv
    -0.82
    enthal
    -0.79
    retty
    -0.77
    arios
    -0.77
    icipated
    -0.76
    ierrez
    -0.72
    yip
    -0.71
    ecided
    -0.70
    POSITIVE LOGITS
    ĪĴ
    0.83
    à¨
    0.75
    äºĶ
    0.63
     stagnation
    0.60
    ®
    0.58
     deaf
    0.58
    ²¾
    0.57
     thor
    0.57
    ×Ļ×
    0.57
    å¦
    0.56
    Activations Density 0.214%

    No Known Activations