© 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. 37340
    Prev
    Next
    INDEX
    Explanations

    terms related to social justice and advocacy for marginalized groups

    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
    314
    +0.21
    1.0%
    160
    +0.16
    0.8%
    442
    +0.13
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    675
    +0.21
    0.07
    650
    +0.16
    0.07
    160
    +0.13
    0.07
    Negative Logits
     Polk
    -0.84
     Caleb
    -0.79
     Mirage
    -0.78
     Villa
    -0.78
     Xiao
    -0.77
     Rez
    -0.76
     Meter
    -0.75
     Mes
    -0.74
     Axel
    -0.73
     Xiang
    -0.73
    POSITIVE LOGITS
    nob
    1.23
    every
    1.21
    null
    1.21
    prep
    1.21
    ever
    1.19
    other
    1.19
    sure
    1.18
    average
    1.16
    new
    1.15
    subject
    1.15
    Activations Density 0.103%

    No Known Activations