© 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. 10-RES-JB
    11. 18522
    Prev
    Next
    INDEX
    Explanations

    words related to the location "Baltimore"

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "Coim" and its variations, often associated with a specific context or location

    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
     loophole
    -0.65
     Pog
    -0.63
     loopholes
    -0.63
     defect
    -0.63
    ¥µ
    -0.60
    scripts
    -0.60
    VALUE
    -0.59
    webkit
    -0.57
     Brah
    -0.56
    aspx
    -0.56
    POSITIVE LOGITS
    rency
    1.32
    incial
    0.91
    idence
    0.88
    identally
    0.88
    cious
    0.87
    ition
    0.85
    inates
    0.83
    atorial
    0.83
    ortium
    0.76
    igion
    0.74
    Activations Density 0.046%

    No Known Activations