© 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. 4-RES-JB
    11. 24162
    Prev
    Next
    INDEX
    Explanations

    phrases indicating violations of rules or laws

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "violate" and its variations

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

    No Comments

    Negative Logits
    arger
    -0.74
    affer
    -0.72
    azor
    -0.69
    anka
    -0.68
    mad
    -0.68
    onna
    -0.67
    agra
    -0.65
    ael
    -0.65
    prototype
    -0.64
    tall
    -0.63
    POSITIVE LOGITS
     unfocusedRange
    0.90
     violations
    0.86
     violation
    0.83
    viol
    0.81
     violating
    0.77
     violates
    0.73
     violate
    0.67
     compliance
    0.67
     violated
    0.67
     breaching
    0.64
    Activations Density 0.018%

    No Known Activations