© 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 · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 9-RES-JB
    11. 10036
    Prev
    Next
    INDEX
    Explanations

    contractions of the phrase "do not"

    oai_token-act-pair · gpt-3.5-turbo

    negative contractions and related phrases indicating refusal or non-compliance

    oai_token-act-pair · gpt-4o-miniTriggered 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
    Not in Any Lists

    No Comments

    Negative Logits
     Published
    -0.67
     Unc
    -0.64
     Cance
    -0.63
     newcom
    -0.58
     parting
    -0.57
     distraction
    -0.56
     doomed
    -0.55
     couch
    -0.55
     caster
    -0.55
     nearest
    -0.55
    POSITIVE LOGITS
    't
    1.43
    ned
    0.98
    uts
    0.95
    eness
    0.90
    keys
    0.90
    ÃŃ
    0.90
    gered
    0.87
    itely
    0.87
    acio
    0.86
    its
    0.83
    Activations Density 0.104%

    No Known Activations