© 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
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 3-RES-JB
    6. 434
    Prev
    Next
    INDEX
    Explanations

    references to legal terms or criminal activities

    oai_token-act-pair · gpt-3.5-turbo

    terms related to legality, specifically focusing on unlawful activities and their contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Tycoon
    -0.78
    acea
    -0.76
    stanbul
    -0.76
    olog
    -0.71
    ingo
    -0.70
    chell
    -0.69
    è¦ļéĨĴ
    -0.69
    issues
    -0.67
    hetti
    -0.67
    oba
    -0.66
    POSITIVE LOGITS
    theless
    0.88
     unfocusedRange
    0.85
     unlawful
    0.84
     Liberties
    0.80
    rontal
    0.75
    00200000
    0.74
     unlawfully
    0.73
     disobedience
    0.71
     disorderly
    0.71
     interference
    0.70
    Activations Density 0.016%

    No Known Activations