© 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. 6658
    Prev
    Next
    INDEX
    Explanations

    words or phrases related to potential risks or hazards

    oai_token-act-pair · gpt-3.5-turbo

    references to various types of dangers or risks

    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
    onement
    -0.88
    BN
    -0.77
    owned
    -0.77
    ergy
    -0.77
    eb
    -0.76
    med
    -0.75
    Ķ
    -0.75
    ced
    -0.71
    oning
    -0.71
    eric
    -0.71
    POSITIVE LOGITS
     pitfalls
    1.10
     dangers
    1.05
     undermin
    1.01
    afety
    0.97
     hazards
    0.91
     risks
    0.81
    ashtra
    0.81
    avascript
    0.78
     endanger
    0.76
     guiActiveUn
    0.76
    Activations Density 0.007%

    No Known Activations