© 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 3072
    9. 8-RES_FS3072-JB
    11. 2992
    Prev
    Next
    INDEX
    Explanations

    phrases related to political events or conflicts

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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
    140
    +0.11
    0.5%
    435
    +0.11
    0.5%
    763
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    541
    +0.11
    0.05
    53
    +0.11
    0.06
    689
    +0.09
    0.05
    Negative Logits
    ampa
    -0.71
    entity
    -0.70
    anta
    -0.69
    ondo
    -0.66
    mitter
    -0.63
     Enhancement
    -0.61
     DOI
    -0.60
    oire
    -0.60
    ffect
    -0.60
    ridge
    -0.60
    POSITIVE LOGITS
     termed
    0.78
    kus
    0.75
    forth
    0.70
     sizeof
    0.66
     aptly
    0.64
    XM
    0.62
    é¾
    0.61
     grossly
    0.60
     euphem
    0.60
    Downloadha
    0.60
    Activations Density 1.094%

    No Known Activations