© 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. 11970
    Prev
    Next
    INDEX
    Explanations

    keywords related to physical objects and places

    oai_token-act-pair · gpt-3.5-turbo

    forms of the word "ard."

    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
    */(
    -0.65
     Kut
    -0.64
     pregnancy
    -0.62
     transmission
    -0.61
     nem
    -0.61
     persu
    -0.61
     backdrop
    -0.60
     surface
    -0.59
     partner
    -0.59
     digit
    -0.58
    POSITIVE LOGITS
    ards
    5.03
    ARDS
    3.13
    ard
    2.66
    arding
    2.65
    arded
    2.58
    ARD
    2.47
    ardless
    1.95
    ardo
    1.67
    arde
    1.54
    ardi
    1.38
    Activations Density 0.013%

    No Known Activations