© 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 3072
    9. 8-RES_FS3072-JB
    11. 749
    Prev
    Next
    INDEX
    Explanations

    technical and statistical terms, or terms related to scientific research and analysis

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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
    538
    +0.13
    0.6%
    208
    +0.11
    0.5%
    690
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    208
    +0.13
    0.09
    89
    +0.11
    0.07
    127
    +0.10
    0.06
    Negative Logits
    nda
    -0.73
    NetMessage
    -0.73
     Archives
    -0.72
    pkg
    -0.72
    JO
    -0.71
    andals
    -0.70
    ruits
    -0.69
    raid
    -0.69
     subpoen
    -0.68
    mail
    -0.67
    POSITIVE LOGITS
     coefficient
    1.26
     ratio
    1.25
    elsius
    1.23
     percentile
    1.23
     cutoff
    1.22
    maximum
    1.22
     threshold
    1.20
     calibrated
    1.19
     Celsius
    1.19
     +/-
    1.17
    Activations Density 2.334%

    No Known Activations