© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 1282
    Prev
    Next
    INDEX
    Explanations

    proper nouns referring to the University of California, Berkeley

    oai_token-act-pair · gpt-3.5-turbo

    references to the word "Berkeley" and variations of it

    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
     Pence
    -0.72
    oons
    -0.70
     Avatar
    -0.70
     Indiana
    -0.70
     PlayStation
    -0.69
     Playstation
    -0.69
    ̶
    -0.69
     CGI
    -0.69
     Shake
    -0.68
    ramid
    -0.68
    POSITIVE LOGITS
    Ber
    3.76
     Berger
    2.26
     Ber
    2.09
     ber
    1.60
     Berg
    1.42
    Mil
    1.32
    Bern
    1.17
    BER
    1.16
    Vi
    1.11
     Berkshire
    1.09
    Activations Density 0.031%

    No Known Activations