© 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. 7834
    Prev
    Next
    INDEX
    Explanations

    phrases related to negative events, such as deaths and arrests

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of death or destruction and their impacts

    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
    mie
    -0.69
    issance
    -0.69
    yip
    -0.67
    erity
    -0.66
    anny
    -0.60
     00
    -0.59
     Tire
    -0.59
     ¯
    -0.58
    chief
    -0.56
    roo
    -0.56
    POSITIVE LOGITS
     themselves
    0.97
     outright
    0.86
     respectively
    0.81
     leases
    0.77
     apiece
    0.77
    onyms
    0.74
    osponsors
    0.74
     individually
    0.73
     voluntarily
    0.73
     redund
    0.72
    Activations Density 0.483%

    No Known Activations