© 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. 0-RES-JB
    11. 15062
    Prev
    Next
    INDEX
    Explanations

    references to specific locations, with a focus on countries

    oai_token-act-pair · gpt-3.5-turbo

    mentions of Malaysia

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Walker
    -0.74
    Walker
    -0.74
     Warden
    -0.73
     Wis
    -0.73
    onne
    -0.72
    Pitt
    -0.71
     tru
    -0.69
     iterator
    -0.68
     sugg
    -0.66
    wic
    -0.66
    POSITIVE LOGITS
     Malaysia
    3.87
     Malaysian
    3.36
     Kuala
    2.95
     Malays
    2.83
     Lumpur
    2.03
     Singapore
    1.91
     MH
    1.89
     Indonesia
    1.84
    Mal
    1.81
     Thailand
    1.74
    Activations Density 0.020%

    No Known Activations