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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 41667
    Prev
    Next
    INDEX
    Explanations

    references to women and their roles or experiences

    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
    501
    +0.11
    0.5%
    584
    +0.10
    0.4%
    556
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    501
    +0.11
    0.02
    325
    +0.10
    0.02
    401
    +0.09
    0.02
    Negative Logits
     Dism
    -0.68
    Coun
    -0.68
    Dur
    -0.66
    Thunder
    -0.64
    San
    -0.63
     Rack
    -0.62
    miss
    -0.62
     Westbrook
    -0.62
    GREEN
    -0.62
    MN
    -0.62
    POSITIVE LOGITS
     alike
    1.58
     respectively
    0.87
     aged
    0.80
     combatants
    0.74
     exchanged
    0.74
    agos
    0.72
    opausal
    0.71
     endowed
    0.71
    acht
    0.70
     striped
    0.70
    Activations Density 0.033%

    No Known Activations