© 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. 0-RES-JB
    11. 20174
    Prev
    Next
    INDEX
    Explanations

    mentions of someone being or acting like a fool

    oai_token-act-pair · gpt-3.5-turbo

    references to being foolish or calling someone a fool

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Extensions
    -0.67
    Pod
    -0.66
     Related
    -0.66
     Kes
    -0.64
     Extended
    -0.63
     Elements
    -0.62
     Contin
    -0.61
     Units
    -0.60
     Events
    -0.59
     Intake
    -0.59
    POSITIVE LOGITS
     fool
    3.92
     fools
    2.45
     Fool
    2.17
     fooled
    1.79
     idiot
    1.75
     deceive
    1.59
     foolish
    1.45
     coward
    1.38
     idiots
    1.36
     dece
    1.28
    Activations Density 0.025%

    No Known Activations