© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 12-RES-JB
    11. 2430
    Prev
    Next
    INDEX
    Explanations

    phrases related to storytelling and narratives

    oai_token-act-pair · gpt-3.5-turbo
    Triggered by @bot

    phrases that relate to the concept of "about," particularly in the context of social discussions or issues

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 12-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_post
    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_post
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Negative Logits
    mite
    -0.79
    prus
    -0.75
    farious
    -0.74
    DragonMagazine
    -0.69
    itute
    -0.69
    enium
    -0.68
    Scroll
    -0.68
    ãģ®å®
    -0.68
    ibi
    -0.67
    azines
    -0.67
    POSITIVE LOGITS
     recognizing
    1.23
     respecting
    1.17
     figuring
    1.16
     acknowledging
    1.14
     knowing
    1.11
     trusting
    1.11
     remembering
    1.10
     overcoming
    1.06
     patience
    1.03
     teamwork
    1.03
    Activations Density 0.377%

    No Known Activations