© 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. 26559
    Prev
    Next
    INDEX
    Explanations

    instances of the word "sitting" and its variations

    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
    +0.12
    0.6%
    469
    +0.12
    0.5%
    546
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    469
    +0.12
    0.03
    582
    +0.12
    0.02
    603
    +0.11
    0.02
    Negative Logits
    xual
    -0.73
    iler
    -0.73
    ctr
    -0.68
    selage
    -0.66
    acco
    -0.65
     CLSID
    -0.65
    iling
    -0.64
    hiba
    -0.62
    usable
    -0.62
    visual
    -0.62
    POSITIVE LOGITS
     atop
    1.09
     ducks
    1.01
     comfortably
    0.94
    uate
    0.89
     idle
    0.80
     dormant
    0.79
     une
    0.78
    DOWN
    0.77
     silently
    0.77
     quietly
    0.75
    Activations Density 0.016%

    No Known Activations