© 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. 1-RES-JB
    11. 1334
    Prev
    Next
    INDEX
    Explanations

    terms related to safety concerns or hazardous conditions

    oai_token-act-pair · gpt-3.5-turbo

    references to safety concerns or hazards

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    thood
    -0.94
    ophers
    -0.90
    ership
    -0.89
    soType
    -0.88
    pard
    -0.84
    bernatorial
    -0.83
    ingham
    -0.82
    uther
    -0.81
    cence
    -0.80
    vation
    -0.80
    POSITIVE LOGITS
     unsafe
    1.24
     unnatural
    0.94
     improperly
    0.81
     unhealthy
    0.80
     unreasonable
    0.78
     improper
    0.77
     adolesc
    0.73
     hazardous
    0.72
     abnormal
    0.70
     wastes
    0.70
    Activations Density 0.013%

    No Known Activations