© 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. 1-RES-JB
    11. 21268
    Prev
    Next
    INDEX
    Explanations

    violations of rules or laws

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "violate" and its variations to identify discussions about breaches of rules or laws

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    uther
    -0.81
    prototype
    -0.76
    aws
    -0.72
     Columb
    -0.72
    aver
    -0.71
    ocker
    -0.69
    pring
    -0.68
    ixed
    -0.66
    batch
    -0.66
    PAC
    -0.66
    POSITIVE LOGITS
     violating
    1.16
     violates
    1.09
     violated
    1.08
     violate
    1.08
     violations
    1.05
     violation
    1.02
    viol
    0.99
     infring
    0.81
     breaching
    0.81
     transgress
    0.80
    Activations Density 0.012%

    No Known Activations