© 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. 30
    Prev
    Next
    INDEX
    Explanations

    phrases conveying authority or commands like "Do not be afraid" or "Don't be fooled"

    oai_token-act-pair · gpt-3.5-turbo

    the phrase "don't be" in various contexts

    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
    inally
    -0.77
     Immunity
    -0.71
    eday
    -0.69
    âĦ¢:
    -0.68
    ciation
    -0.66
    albeit
    -0.64
    lux
    -0.64
     flourish
    -0.63
    issance
    -0.63
    offer
    -0.63
    POSITIVE LOGITS
     fooled
    1.19
     deceived
    1.05
     misled
    1.04
     confused
    1.03
     swayed
    1.03
     bothered
    1.01
     afraid
    0.99
     underestimated
    0.98
     intimidated
    0.98
     blamed
    0.94
    Activations Density 0.106%

    No Known Activations