© 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. 10-RES-JB
    11. 10197
    Prev
    Next
    INDEX
    Explanations

    phrases related to unethical behavior or legal investigations, especially those related to corruption, sexual misconduct, bribery, and money laundering

    oai_token-act-pair · gpt-3.5-turbo

    terms related to allegations of misconduct and corruption

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    izont
    -0.76
    icon
    -0.71
    adle
    -0.71
    random
    -0.69
    ethe
    -0.69
    sonian
    -0.69
    liam
    -0.68
    ãĥ¥
    -0.68
    ellar
    -0.67
    itar
    -0.67
    POSITIVE LOGITS
     allegations
    1.37
     disclosures
    1.17
     scandals
    1.14
     violations
    1.11
     accusations
    1.10
     charges
    1.10
     wrongdoing
    1.07
     offences
    1.07
     abuses
    1.00
     misconduct
    1.00
    Activations Density 0.202%

    No Known Activations