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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 13132
    Prev
    Next
    INDEX
    Explanations

    phrases related to fraudulent schemes or conspiracies

    oai_token-act-pair · gpt-3.5-turbo

    terms related to schemes, conspiracies, and manipulation

    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
     aging
    -0.74
     ageing
    -0.72
    odox
    -0.70
    status
    -0.66
     clarity
    -0.65
     Quality
    -0.64
     condol
    -0.64
     Mobility
    -0.64
     disabilities
    -0.63
     reflections
    -0.63
    POSITIVE LOGITS
     perpetrated
    1.49
     orchestrated
    1.40
     hatched
    1.33
     concoct
    1.30
     devised
    1.12
     disguised
    1.09
     scam
    1.08
    Operation
    1.07
     disgu
    1.03
     waged
    1.02
    Activations Density 0.489%

    No Known Activations