© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 9836
    Prev
    Next
    INDEX
    Explanations

    phrases related to putting something on hold or suspension

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating something being placed "on hold" or paused

    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
     Rooms
    -0.72
    omez
    -0.68
    SourceFile
    -0.65
    xs
    -0.64
    enegger
    -0.64
    elf
    -0.60
    Untitled
    -0.60
    orically
    -0.57
    conom
    -0.56
     Incarn
    -0.55
    POSITIVE LOGITS
     hold
    1.08
     pause
    1.06
     pedest
    1.03
     hiatus
    1.00
     display
    0.95
     notice
    0.93
     lockdown
    0.92
     indefinite
    0.91
     cooldown
    0.86
     standby
    0.85
    Activations Density 0.069%

    No Known Activations