© 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 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. 12442
    Prev
    Next
    INDEX
    Explanations

    phrases indicating something has been unfairly manipulated or set up in a specific way

    oai_token-act-pair · gpt-3.5-turbo

    mentions of the term "rigged" in various contexts

    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
     Orig
    -0.71
     Prairie
    -0.70
     visitor
    -0.70
    Pages
    -0.69
     traveller
    -0.66
     Uriel
    -0.65
     Angels
    -0.64
     Psy
    -0.64
     Franz
    -0.62
     Cullen
    -0.61
    POSITIVE LOGITS
     rigged
    4.15
     rigging
    2.80
    rig
    1.38
     rigs
    1.25
     crooked
    1.18
     stacked
    1.14
     sabot
    1.13
     rig
    1.11
     biased
    1.07
     hijacked
    1.07
    Activations Density 0.032%

    No Known Activations