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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 21342
    Prev
    Next
    INDEX
    Explanations

    phrases related to legal guilt or pronouncements of guilt

    oai_token-act-pair · gpt-3.5-turbo

    references to guilt and guilty verdicts

    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
    psey
    -0.81
     Gork
    -0.69
    Secure
    -0.66
    Dub
    -0.64
    LET
    -0.62
    ANN
    -0.61
    atility
    -0.61
    atile
    -0.61
    oper
    -0.61
    quickShipAvailable
    -0.59
    POSITIVE LOGITS
     plea
    1.06
     verdict
    1.01
     pleas
    0.97
     conscience
    0.97
     bystanders
    0.88
     bystand
    0.88
     guilt
    0.83
     pleasures
    0.83
     innocence
    0.82
     innocent
    0.72
    Activations Density 0.083%

    No Known Activations