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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 42943
    Prev
    Next
    INDEX
    Explanations

    casual expressions of limitation or condition

    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
    101
    +0.12
    0.5%
    512
    +0.09
    0.4%
    221
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    221
    +0.12
    0.01
    117
    +0.09
    0.01
    734
    +0.09
    0.01
    Negative Logits
    itas
    -0.69
    inally
    -0.69
    iol
    -0.67
    ulators
    -0.67
     glim
    -0.64
    iture
    -0.64
     reperto
    -0.63
    eport
    -0.63
    igers
    -0.62
     plent
    -0.61
    POSITIVE LOGITS
     whatsoever
    0.97
     disrespect
    0.82
     necessarily
    0.81
     nor
    0.79
     prejud
    0.76
     detract
    0.76
     harmed
    0.75
     condone
    0.72
     guarantees
    0.72
     discouraged
    0.70
    Activations Density 0.016%

    No Known Activations