© 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. 4-RES-JB
    11. 10442
    Prev
    Next
    INDEX
    Explanations

    terms related to controversial topics or issues

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "controversial."

    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
    ynthesis
    -0.86
    vation
    -0.81
    amina
    -0.76
    igi
    -0.74
    á
    -0.74
    xia
    -0.71
    anza
    -0.71
    amen
    -0.70
    elsen
    -0.69
    ©¶æ
    -0.69
    POSITIVE LOGITS
     controversial
    0.89
     fringe
    0.77
     topics
    0.75
     controversies
    0.75
     unpopular
    0.73
     taboo
    0.71
     outl
    0.70
     topic
    0.70
     aspects
    0.69
     opinions
    0.69
    Activations Density 0.018%

    No Known Activations