© 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. 8-RES-JB
    11. 7984
    Prev
    Next
    INDEX
    Explanations

    profane language and derogatory terms

    oai_token-act-pair · gpt-3.5-turbo

    phrases that express strong opinions or controversial statements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    atories
    -0.66
     evolves
    -0.60
    UTC
    -0.60
     bilateral
    -0.57
     intangible
    -0.55
    ļéĨĴ
    -0.54
     progresses
    -0.54
    OTOS
    -0.54
    gemony
    -0.52
     bruising
    -0.52
    POSITIVE LOGITS
    joice
    0.69
    haven
    0.68
    Well
    0.67
     rejoice
    0.65
    may
    0.65
    joy
    0.65
    agine
    0.65
     doubtless
    0.65
     surely
    0.64
    utm
    0.62
    Activations Density 0.384%

    No Known Activations