© 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
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 0-RES-JB
    6. 4975
    Prev
    Next
    INDEX
    Explanations

    chemical compounds or substances that are classified as carcinogenic

    oai_token-act-pair · gpt-3.5-turbo

    terms related to cancer-causing substances and 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
    CHAR
    -0.68
     Blvd
    -0.63
     subt
    -0.63
     threads
    -0.63
     scraps
    -0.62
     shirt
    -0.62
     fence
    -0.62
     guides
    -0.62
     sheriff
    -0.61
     fences
    -0.61
    POSITIVE LOGITS
    ogenic
    4.35
    ogens
    2.74
    ogenesis
    2.48
    ogen
    2.33
    ogene
    1.61
    ophys
    1.33
    ologic
    1.30
    ochemical
    1.26
    otropic
    1.26
    omorphic
    1.26
    Activations Density 0.016%

    No Known Activations