© 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. 10-RES-JB
    11. 11128
    Prev
    Next
    INDEX
    Explanations

    phrases related to legal actions or allegations

    oai_token-act-pair · gpt-3.5-turbo

    references to legal actions and allegations

    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
    haar
    -0.77
    tan
    -0.74
    wait
    -0.73
    PsyNetMessage
    -0.73
    cffffcc
    -0.72
    vantage
    -0.71
    milo
    -0.71
    think
    -0.70
    toggle
    -0.69
    population
    -0.68
    POSITIVE LOGITS
     negligence
    1.08
     wrongdoing
    1.08
     violations
    1.06
     wrongful
    0.98
     irregularities
    0.95
     improper
    0.93
     fraud
    0.90
     that
    0.90
     misconduct
    0.90
     breach
    0.89
    Activations Density 0.110%

    No Known Activations