© 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. 7763
    Prev
    Next
    INDEX
    Explanations

    phrases relating to being unharmed or safe

    oai_token-act-pair · gpt-3.5-turbo

    mentions of the word "har" or related variations, possibly indicating a focus on the term's usage in various contexts

    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
    htaking
    -0.91
    lect
    -0.81
    essee
    -0.78
    uring
    -0.70
    hift
    -0.70
    BOOK
    -0.70
    urally
    -0.69
    olve
    -0.67
     fuzz
    -0.64
     chalk
    -0.64
    POSITIVE LOGITS
    assment
    1.01
    tha
    0.99
    allel
    0.93
     Tsarnaev
    0.91
    vard
    0.90
    riors
    0.89
    riage
    0.85
    riages
    0.84
    ashtra
    0.82
    adow
    0.80
    Activations Density 0.066%

    No Known Activations