© 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 EvalsExportsinterp-engineNEWAPI 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. 12378
    Prev
    Next
    INDEX
    Explanations

    comparative statements or ranking labels indicated by the use of hyphens and numbers

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating order or ranking, particularly in the context of drafts and selections

    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
    Ń·
    -0.68
     mars
    -0.64
     PD
    -0.61
     formula
    -0.61
     Robertson
    -0.61
    =-=-=-=-
    -0.60
     insecure
    -0.60
     Tant
    -0.58
     pul
    -0.57
    âĸĵ
    -0.57
    POSITIVE LOGITS
    generation
    1.38
    ever
    1.33
    degree
    1.29
    timer
    1.29
    party
    1.22
    rate
    1.21
    person
    1.20
    class
    1.20
    round
    1.18
    edition
    1.17
    Activations Density 0.022%

    No Known Activations