© 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. 14105
    Prev
    Next
    INDEX
    Explanations

    phrases related to exaggeration or subtlety

    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
    180
    +0.12
    0.6%
    274
    +0.11
    0.5%
    729
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    274
    +0.12
    0.05
    729
    +0.11
    0.04
    737
    +0.11
    0.04
    Negative Logits
    roots
    -0.72
     exting
    -0.66
    Untitled
    -0.63
    emis
    -0.63
    ourses
    -0.63
     renovated
    -0.63
    otypes
    -0.61
    ãĥĪ
    -0.60
     Citiz
    -0.60
    rete
    -0.60
    POSITIVE LOGITS
     understatement
    0.97
     haha
    0.93
     considering
    0.89
     assumption
    0.85
     ;)
    0.85
     :-)
    0.82
     conjecture
    0.82
     :)
    0.81
     proposition
    0.80
     exaggeration
    0.80
    Activations Density 0.298%

    No Known Activations