© 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. 9-RES-JB
    11. 24053
    Prev
    Next
    INDEX
    Explanations

    phrases related to intense or negative emotions

    oai_token-act-pair · gpt-3.5-turbo

    negative emotions or experiences often expressed through idioms or metaphors

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 9-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    9
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Negative Logits
     Compass
    -0.63
     externalToEVAOnly
    -0.63
     fixme
    -0.62
     showc
    -0.62
    ĨĴ
    -0.61
    uration
    -0.61
     hospitality
    -0.61
     accommodation
    -0.59
    ;;;;;;;;;;;;
    -0.59
    etus
    -0.59
    POSITIVE LOGITS
    shit
    1.09
    dash
    1.00
    faced
    0.95
    buck
    0.94
    footed
    0.91
    haired
    0.89
    breaker
    0.89
    ball
    0.88
    hearted
    0.86
    puff
    0.86
    Activations Density 0.096%

    No Known Activations