© 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 EvalsExportsinterp-engineNEWAPI 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. 0-RES-JB
    11. 8165
    Prev
    Next
    INDEX
    Explanations

    verbs related to circumventing or bypassing something

    oai_token-act-pair · gpt-3.5-turbo

    terms related to evading rules or processes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    eared
    -0.82
    astic
    -0.77
    imony
    -0.76
    Orig
    -0.75
    atl
    -0.70
    ears
    -0.69
    ust
    -0.65
     Soci
    -0.64
    ear
    -0.64
    arling
    -0.63
    POSITIVE LOGITS
     bypass
    3.82
     circumvent
    2.50
     overr
    1.17
     skip
    1.16
     sid
    1.15
     override
    1.13
     workaround
    1.04
     skipping
    1.04
     skipped
    1.04
     travers
    1.03
    Activations Density 0.014%

    No Known Activations