© 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 EvalsExportsinterp-engineNEWAPI 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. 14001
    Prev
    Next
    INDEX
    Explanations

    countries and regions with associated effects or impacts

    oai_token-act-pair · gpt-3.5-turbo

    key terms related to social and economic impacts

    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
    ITNESS
    -0.70
     Birthday
    -0.64
     Canaver
    -0.61
    *)
    -0.61
    edIn
    -0.59
     consisted
    -0.57
     consists
    -0.57
    ijah
    -0.56
    ©¶æ
    -0.55
    htaking
    -0.54
    POSITIVE LOGITS
     unnecessarily
    1.05
     thereby
    0.98
     tremendously
    0.97
     considerably
    0.96
     altogether
    0.95
     quicker
    0.94
     enormously
    0.93
     faster
    0.90
     by
    0.90
     immensely
    0.90
    Activations Density 0.500%

    No Known Activations