© 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 49152
    9. 8-RES_FS49152-JB
    11. 24013
    Prev
    Next
    INDEX
    Explanations

    numerical ratings or evaluations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    430
    +0.12
    0.6%
    506
    +0.10
    0.5%
    218
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    303
    +0.12
    0.03
    632
    +0.10
    0.03
    335
    +0.10
    0.03
    Negative Logits
     capitals
    -0.73
     advis
    -0.70
     roam
    -0.67
     stewards
    -0.67
     budgets
    -0.66
    ussion
    -0.65
    etheless
    -0.65
     leap
    -0.64
     purse
    -0.63
     dotted
    -0.63
    POSITIVE LOGITS
    J
    1.03
    E
    0.97
    O
    0.95
    A
    0.91
    Y
    0.90
    Va
    0.86
    C
    0.85
    K
    0.84
    L
    0.83
    D
    0.83
    Activations Density 0.031%

    No Known Activations