© 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. 3352
    Prev
    Next
    INDEX
    Explanations

    verbs in the form of 'have' followed by an associated action or state

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "have."

    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
     Recap
    -0.68
    /+
    -0.65
    uce
    -0.64
    ove
    -0.63
    oyer
    -0.62
    sea
    -0.61
    OF
    -0.61
     Present
    -0.60
     Merchant
    -0.59
    inks
    -0.58
    POSITIVE LOGITS
     been
    1.28
    been
    1.09
     begun
    1.07
     reacted
    1.06
     gotten
    1.05
     complained
    1.03
     recourse
    1.00
     benefited
    0.96
     endured
    0.96
     resorted
    0.93
    Activations Density 0.218%

    No Known Activations