© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 3-RES-JB
    11. 17000
    Prev
    Next
    INDEX
    Explanations

    words related to triggers or activation mechanisms

    oai_token-act-pair · gpt-3.5-turbo

    the repeated occurrence and variation of the word "trigger" in various contexts

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Negative Logits
    hemat
    -0.81
    ļéĨĴ
    -0.71
     Partnership
    -0.69
    aredevil
    -0.68
     Expend
    -0.67
     Cutler
    -0.63
    ibble
    -0.63
    ī
    -0.62
    phis
    -0.62
     Accounting
    -0.62
    POSITIVE LOGITS
     trigger
    1.12
    trigger
    1.11
     triggers
    1.10
     triggering
    1.01
     Trigger
    0.85
    uating
    0.82
     warnings
    0.79
    fires
    0.75
     derail
    0.75
     triggered
    0.75
    Activations Density 0.009%

    No Known Activations