© 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. 8-RES-JB
    11. 14138
    Prev
    Next
    INDEX
    Explanations

    action verbs followed by prepositions, particularly 'to'

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating partnerships and collaborations

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Negative Logits
    quickShipAvailable
    -0.79
    thing
    -0.71
    ults
    -0.66
    erence
    -0.63
    Jess
    -0.61
    Joy
    -0.61
    anders
    -0.61
    Winged
    -0.61
     glances
    -0.60
     wat
    -0.59
    POSITIVE LOGITS
     ensure
    1.28
     create
    1.23
     deliver
    1.14
     develop
    1.13
     formulate
    1.10
     provide
    1.09
     produce
    1.08
     promote
    1.07
     recreate
    1.05
     maximize
    1.05
    Activations Density 0.172%

    No Known Activations