© 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. 3-RES-JB
    11. 507
    Prev
    Next
    INDEX
    Explanations

    mentions of the word "collusion" and related terms like "complicity."

    oai_token-act-pair · gpt-3.5-turbo

    references to collusion and complicity in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Hop
    -0.84
     Balanced
    -0.74
    =-=-=-=-=-=-=-=-
    -0.73
    ĻĤ
    -0.70
     Rite
    -0.69
    zl
    -0.69
    thur
    -0.67
    mac
    -0.67
    ulton
    -0.67
     Proud
    -0.67
    POSITIVE LOGITS
     collusion
    1.20
     complicity
    0.95
     complicit
    0.94
     conspiring
    0.87
     tampering
    0.85
     unfocusedRange
    0.84
     wink
    0.79
     blacklist
    0.78
     nexus
    0.77
     exoner
    0.72
    Activations Density 0.010%

    No Known Activations