© 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. 4-RES-JB
    11. 11466
    Prev
    Next
    INDEX
    Explanations

    the word "how" in different contexts and variations

    oai_token-act-pair · gpt-3.5-turbo

    expressions related to the concept of "how" and temporal or contextual references

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

    No Comments

    Negative Logits
    ĸļ
    -1.13
    onde
    -0.71
    ICLE
    -0.68
    netflix
    -0.68
    =-=-=-=-=-=-=-=-
    -0.67
     Debor
    -0.66
    enegger
    -0.66
     Dunham
    -0.65
    anski
    -0.61
    Fram
    -0.61
    POSITIVE LOGITS
    istically
    0.71
    else
    0.70
     stranger
    0.70
    worldly
    0.70
    istical
    0.67
    place
    0.66
     somew
    0.64
     flair
    0.63
    away
    0.63
     wiser
    0.63
    Activations Density 0.158%

    No Known Activations