© 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. 2-RES-JB
    11. 433
    Prev
    Next
    INDEX
    Explanations

    the word "what"

    oai_token-act-pair · gpt-4-turbo

    the pronoun "what" followed by a verb or a noun

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "what."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ster
    -0.67
    trop
    -0.67
    fish
    -0.66
    robe
    -0.66
    uttering
    -0.63
    ped
    -0.62
    Gy
    -0.61
    por
    -0.61
    eer
    -0.59
    ohyd
    -0.59
    POSITIVE LOGITS
    soever
    1.24
     happens
    1.09
     happened
    1.06
     sorts
    1.00
     happ
    0.98
     kinds
    0.97
     transpired
    0.91
     exactly
    0.84
     else
    0.79
     constitutes
    0.78
    Activations Density 0.141%

    No Known Activations