© 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 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. 0-RES-JB
    11. 9669
    Prev
    Next
    INDEX
    Explanations

    proper nouns related to characters or locations in different contexts

    oai_token-act-pair · gpt-3.5-turbo

    proper nouns, particularly names of characters, places, and significant entities

    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
    Rub
    -0.78
    bid
    -0.73
     jug
    -0.70
    inning
    -0.70
    iliate
    -0.69
     Beng
    -0.69
    edient
    -0.68
     Hos
    -0.67
    fan
    -0.67
    Yan
    -0.67
    POSITIVE LOGITS
     Normandy
    4.08
     Shepard
    1.67
     Calais
    1.60
     Cerberus
    1.51
     Sicily
    1.39
     Lancaster
    1.33
     Cologne
    1.32
     Brittany
    1.28
     Canterbury
    1.26
     Dresden
    1.23
    Activations Density 0.060%

    No Known Activations