© 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. 10-RES-JB
    11. 13972
    Prev
    Next
    INDEX
    Explanations

    names of locations or entities related to a particular program or project, possibly involving contractors and energy upgrades

    oai_token-act-pair · gpt-3.5-turbo

    expressions of welcome or invitation

    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
     Trinidad
    -0.69
     sunset
    -0.66
     harmon
    -0.66
     friction
    -0.65
     dusk
    -0.64
     Rhodes
    -0.63
     Laurent
    -0.63
     Wilmington
    -0.63
     unification
    -0.62
     Coliseum
    -0.61
    POSITIVE LOGITS
    must
    1.21
    should
    1.12
    ll
    1.11
    need
    1.07
    ve
    1.06
    swer
    1.06
    know
    1.05
    want
    1.04
    re
    1.03
    ï¸ı
    1.03
    Activations Density 0.124%

    No Known Activations