© 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. 22558
    Prev
    Next
    INDEX
    Explanations

    discussions related to government legislation and bills

    oai_token-act-pair · gpt-4o-miniTriggered 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
    129
    +0.10
    0.4%
    206
    +0.09
    0.4%
    48
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    680
    +0.10
    0.04
    620
    +0.09
    0.02
    300
    +0.09
    0.06
    Negative Logits
    $$$$
    -0.67
    ertodd
    -0.64
    uana
    -0.63
    amoto
    -0.62
    Production
    -0.61
    ...]
    -0.60
    minist
    -0.60
     PLEASE
    -0.60
     groove
    -0.59
    bt
    -0.59
    POSITIVE LOGITS
     previous
    1.09
     unsuccessfully
    1.05
     earlier
    1.01
     predecessor
    0.94
     similarly
    0.91
     predecessors
    0.91
     prior
    0.89
     unsuccessful
    0.88
     similar
    0.86
     preceded
    0.84
    Activations Density 0.863%

    No Known Activations