© 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. 0-RES-JB
    11. 12958
    Prev
    Next
    INDEX
    Explanations

    words ending in '-ching' with high activation values

    oai_token-act-pair · gpt-3.5-turbo

    gerunds and present participles, particularly those ending in 'ching'

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     VAL
    -0.73
     interpret
    -0.68
     intellig
    -0.62
     credit
    -0.62
     analy
    -0.62
     value
    -0.61
     counsel
    -0.61
     Interpret
    -0.60
     interpreted
    -0.60
     Flav
    -0.60
    POSITIVE LOGITS
    ching
    4.69
    ched
    2.80
    ches
    2.50
    chy
    2.00
    chers
    1.87
    cher
    1.77
    ch
    1.63
    chery
    1.55
    ting
    1.30
    king
    1.23
    Activations Density 0.010%

    No Known Activations