© 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. 1114
    Prev
    Next
    INDEX
    Explanations

    phrases indicating exceptions, contradictions, or qualifications

    oai_token-act-pair · gpt-3.5-turbo

    phrases that emphasize negation or the absence of something

    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
     Quote
    -0.67
     respectively
    -0.66
    ]}
    -0.65
     whichever
    -0.62
     awoken
    -0.62
    GBT
    -0.62
    pta
    -0.61
    anwhile
    -0.59
     Countdown
    -0.58
    aukee
    -0.58
    POSITIVE LOGITS
     means
    1.07
     coincidence
    0.87
     stretch
    0.85
     accident
    0.77
     avail
    0.75
     Means
    0.72
     exaggeration
    0.72
    product
    0.72
     doubt
    0.71
     fault
    0.70
    Activations Density 0.038%

    No Known Activations