© 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 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. 10-RES-JB
    11. 16563
    Prev
    Next
    INDEX
    Explanations

    mentions of the word "Merck" with varying degrees of misspelling or truncation

    oai_token-act-pair · gpt-3.5-turbo

    references to the pharmaceutical company Merck

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     numbered
    -0.61
    IGHTS
    -0.60
     sidelines
    -0.59
     fake
    -0.59
    ¬¼
    -0.57
     Scarborough
    -0.57
     hunger
    -0.57
     smoker
    -0.57
     hanging
    -0.57
     Everywhere
    -0.57
    POSITIVE LOGITS
    cedes
    1.22
    seys
    1.16
    lins
    1.13
    rily
    1.12
    chant
    1.11
    maid
    1.07
    itage
    1.06
    ging
    1.02
    gency
    1.00
    rim
    1.00
    Activations Density 0.015%

    No Known Activations