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

    Neuronpedia

    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. 37411
    Prev
    Next
    INDEX
    Explanations

    phrases expressing negation or refusal

    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
    437
    +0.13
    0.6%
    665
    +0.12
    0.5%
    447
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    665
    +0.13
    0.04
    645
    +0.12
    0.03
    437
    +0.10
    0.03
    Negative Logits
    eyes
    -0.78
    quickShipAvailable
    -0.76
    facts
    -0.70
    ãĤ¼ãĤ¦ãĤ¹
    -0.68
    eers
    -0.66
     dynamics
    -0.65
    tips
    -0.65
    lies
    -0.63
    SER
    -0.63
    icons
    -0.61
    POSITIVE LOGITS
     approve
    1.00
     pursue
    1.00
     comply
    1.00
     recommend
    0.96
     renew
    0.91
     endorse
    0.87
     prosecute
    0.86
     vacc
    0.86
    icably
    0.85
     celebrate
    0.84
    Activations Density 0.067%

    No Known Activations