© 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. 9136
    Prev
    Next
    INDEX
    Explanations

    personal pronouns

    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
    480
    +0.32
    1.6%
    447
    +0.17
    0.9%
    566
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    502
    +0.32
    0.08
    566
    +0.17
    0.08
    93
    +0.13
    0.07
    Negative Logits
    minster
    -0.65
     whichever
    -0.56
     Bever
    -0.55
     Malone
    -0.54
     totality
    -0.53
     antid
    -0.52
     Impact
    -0.52
    illary
    -0.51
     conservation
    -0.51
     Gad
    -0.51
    POSITIVE LOGITS
    'm
    1.37
    've
    1.20
     suppose
    0.99
    WI
    0.99
    'll
    0.98
    'd
    0.98
     am
    0.96
    EEE
    0.94
    ANS
    0.91
    deals
    0.91
    Activations Density 0.181%

    No Known Activations