© 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. 11-RES-JB
    11. 21804
    Prev
    Next
    INDEX
    Explanations

    phrases related to substances that are harmful or may cause health issues

    oai_token-act-pair · gpt-3.5-turbo

    terms related to health, safety, and potential hazards in products or substances

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    nesday
    -0.85
    ruary
    -0.81
    opolis
    -0.73
    ridor
    -0.71
    ISION
    -0.70
    essage
    -0.70
    antage
    -0.69
    Particip
    -0.68
    Lv
    -0.67
    pee
    -0.67
    POSITIVE LOGITS
     fungi
    1.64
     pathogens
    1.53
     viruses
    1.53
     parasites
    1.52
     toxins
    1.52
     pesticides
    1.45
     enzymes
    1.44
     allerg
    1.41
     pollutants
    1.40
     microbes
    1.39
    Activations Density 0.222%

    No Known Activations