© 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. 11-RES-JB
    11. 9128
    Prev
    Next
    INDEX
    Explanations

    adverbs ending in '-ly'

    oai_token-act-pair · gpt-3.5-turbo

    adverbs that describe manner or frequency in actions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     ado
    -0.67
    ilater
    -0.61
    ãĥ¯ãĥ³
    -0.61
    eries
    -0.61
    rite
    -0.60
     totally
    -0.60
    enf
    -0.60
     privilege
    -0.58
     coincidence
    -0.57
     hemor
    -0.57
    POSITIVE LOGITS
     enough
    1.11
    enough
    1.04
     throughout
    0.90
     thereafter
    0.82
    tics
    0.81
     through
    0.79
     during
    0.77
     across
    0.77
     speaking
    0.75
     whenever
    0.73
    Activations Density 0.161%

    No Known Activations