© 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. 41661
    Prev
    Next
    INDEX
    Explanations

    words that describe composition or elements of items

    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
    679
    +0.17
    0.8%
    403
    +0.10
    0.5%
    742
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    403
    +0.17
    0.01
    679
    +0.10
    0.01
    541
    +0.10
    0.01
    Negative Logits
    ilings
    -0.65
    hawks
    -0.64
    fly
    -0.61
    antha
    -0.61
    sell
    -0.60
     Karma
    -0.58
    Hill
    -0.58
    abad
    -0.58
     Pilgrim
    -0.57
    zee
    -0.56
    POSITIVE LOGITS
     solely
    0.90
     principally
    0.86
     chiefly
    0.86
     mainly
    0.85
     primarily
    0.80
    antly
    0.78
    rences
    0.76
    encies
    0.76
     prominently
    0.75
     entirely
    0.74
    Activations Density 0.015%

    No Known Activations