© 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. 25763
    Prev
    Next
    INDEX
    Explanations

    phrases or terms indicating fundamental or root causes of issues

    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
    99
    +0.11
    0.5%
    195
    +0.11
    0.5%
    288
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    99
    +0.11
    0.01
    455
    +0.11
    0.01
    78
    +0.10
    0.01
    Negative Logits
    ishers
    -0.83
    UNCH
    -0.82
    ooters
    -0.82
    cture
    -0.82
    uden
    -0.81
    hops
    -0.81
    isters
    -0.81
    asia
    -0.78
    aterasu
    -0.78
    alde
    -0.78
    POSITIVE LOGITS
     assumption
    1.08
     structure
    1.04
     assumptions
    1.04
     principles
    1.02
     fundamentals
    1.02
     underlying
    1.01
     theme
    1.01
     premise
    1.00
     motivations
    1.00
     infrastructure
    0.99
    Activations Density 0.011%

    No Known Activations