© 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. 4-RES-JB
    11. 7534
    Prev
    Next
    INDEX
    Explanations

    references to geographical locations, specifically focusing on the city of Manila and the country of the Philippines

    oai_token-act-pair · gpt-3.5-turbo

    references to the locations and entities related to the Philippines, particularly Manila

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

    No Comments

    Negative Logits
    schild
    -0.89
    Ö¼
    -0.79
    ø
    -0.78
    ãĥ¤
    -0.77
    heimer
    -0.76
    ership
    -0.75
    chool
    -0.74
    ãĥł
    -0.74
     Macintosh
    -0.73
    ivities
    -0.72
    POSITIVE LOGITS
     Manila
    1.06
     Filip
    0.93
     Philippines
    0.92
     Islands
    0.89
     pes
    0.86
     typh
    0.86
     Airlines
    0.78
     Duterte
    0.77
     Filipino
    0.75
     Rodrigo
    0.74
    Activations Density 0.020%

    No Known Activations