© 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. 11-RES-JB
    11. 5194
    Prev
    Next
    INDEX
    Explanations

    mentions of guns, gun-related terms, and related consequences such as deaths

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    adem
    -0.61
     most
    -0.60
    olicy
    -0.59
    escription
    -0.59
    ĺħ
    -0.57
    best
    -0.57
     best
    -0.56
    «ĺ
    -0.56
     Lazarus
    -0.55
     entirety
    -0.55
    POSITIVE LOGITS
     than
    2.00
    than
    1.56
     Than
    1.56
     equals
    0.79
    efficient
    0.78
    worldly
    0.78
     per
    0.69
    rah
    0.68
    antry
    0.67
    iable
    0.66
    Activations Density 0.329%

    No Known Activations