© 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. 10-RES-JB
    11. 18962
    Prev
    Next
    INDEX
    Explanations

    historical events and political actions related to annexation or conquest

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to territorial control and historical governance

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    wcsstore
    -0.89
    inventoryQuantity
    -0.83
    BILL
    -0.77
    quickShipAvailable
    -0.77
     motivational
    -0.75
    advertisement
    -0.71
     solicitation
    -0.69
     sniper
    -0.68
    nikov
    -0.68
    DragonMagazine
    -0.67
    POSITIVE LOGITS
     Europe
    1.70
     Britain
    1.47
     Ireland
    1.41
     Persia
    1.40
     Poland
    1.38
     Yugoslavia
    1.37
     France
    1.35
     England
    1.35
     India
    1.35
     Asia
    1.34
    Activations Density 0.353%

    No Known Activations