© 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. 2-RES-JB
    11. 14026
    Prev
    Next
    INDEX
    Explanations

    mentions of the word "anthrax"

    oai_token-act-pair · gpt-3.5-turbo

    references to anthrax

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Aad
    -0.69
    ================================================================
    -0.65
    ĨĴ
    -0.64
     NFC
    -0.63
     Marlins
    -0.61
    ä
    -0.60
     Carth
    -0.60
    mable
    -0.59
    Compat
    -0.59
     Holden
    -0.59
    POSITIVE LOGITS
    rax
    1.46
    itect
    0.98
    eus
    0.90
    ionage
    0.88
    ounding
    0.84
    inus
    0.81
    illin
    0.78
    irtual
    0.77
    enses
    0.77
    iar
    0.77
    Activations Density 0.013%

    No Known Activations