© 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. 10-RES-JB
    11. 18553
    Prev
    Next
    INDEX
    Explanations

    the prefix "un-" followed by an action or state in the text

    oai_token-act-pair · gpt-3.5-turbo

    words or prefixes related to the concept of "un-," indicating negation or reversal

    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
     watered
    -0.68
     unpop
    -0.63
     recurring
    -0.63
     bedrooms
    -0.63
     Simpson
    -0.62
     Tut
    -0.61
     untold
    -0.61
     chase
    -0.60
     unexplained
    -0.59
     home
    -0.59
    POSITIVE LOGITS
    rave
    1.26
    plug
    1.17
    loading
    1.11
    cles
    1.11
    offic
    1.10
    load
    1.10
    apolog
    1.09
    zip
    1.05
    ifies
    1.04
    seat
    1.04
    Activations Density 0.022%

    No Known Activations