© 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. 26671
    Prev
    Next
    INDEX
    Explanations

    possessive pronouns and their corresponding subjects in various contexts

    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
    481
    +0.17
    0.8%
    175
    +0.12
    0.5%
    553
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.17
    0.07
    553
    +0.12
    0.05
    15
    +0.11
    0.06
    Negative Logits
    ghazi
    -0.99
    tenance
    -0.81
    security
    -0.81
    lehem
    -0.80
    justice
    -0.80
    Security
    -0.78
    hari
    -0.78
    needed
    -0.77
    gate
    -0.77
    payers
    -0.76
    POSITIVE LOGITS
     masterpiece
    1.29
     own
    1.24
     autobiography
    1.17
     namesake
    1.13
     creations
    1.13
     seminal
    1.07
     compositions
    1.06
     signature
    1.05
     trademark
    1.05
     newest
    1.04
    Activations Density 0.210%

    No Known Activations