© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    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. 14730
    Prev
    Next
    INDEX
    Explanations

    locations where specific events occurred

    oai_token-act-pair · gpt-3.5-turbo

    references to specific locations related to incidents or events

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    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
    language
    -0.69
     caveat
    -0.68
    eness
    -0.68
     notwithstanding
    -0.68
    ragon
    -0.68
    medium
    -0.67
    annel
    -0.67
    grim
    -0.66
     spite
    -0.66
    unning
    -0.65
    POSITIVE LOGITS
     Trayvon
    0.81
     gunmen
    0.78
     slain
    0.77
     murdered
    0.76
     they
    0.73
     Osama
    0.72
     robbers
    0.71
     gunman
    0.71
     Paddock
    0.71
     abducted
    0.70
    Activations Density 0.203%

    No Known Activations