© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 9755
    Prev
    Next
    INDEX
    Explanations

    measurements or assessments related to different aspects

    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
    481
    +0.15
    0.7%
    127
    +0.12
    0.5%
    557
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.15
    0.04
    97
    +0.12
    0.03
    608
    +0.09
    0.04
    Negative Logits
     Carbuncle
    -0.77
    irez
    -0.76
     Canaver
    -0.76
    Synopsis
    -0.76
    arak
    -0.76
    rim
    -0.74
    emi
    -0.72
    angs
    -0.71
    mare
    -0.71
    soever
    -0.71
    POSITIVE LOGITS
     effectiveness
    1.34
     probabilities
    1.28
     probability
    1.26
     prevalence
    1.19
     accuracy
    1.18
     attractiveness
    1.18
     popularity
    1.18
     likelihood
    1.16
     efficacy
    1.16
     usefulness
    1.13
    Activations Density 0.215%

    No Known Activations