Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 0-RES-JB
    6. 3574
    Prev
    Next
    INDEX
    Explanations

    specific words related to a particular beer festival called "Borefts"

    oai_token-act-pair · gpt-3.5-turbo

    mentions of specific breweries and their associated events

    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
     Requ
    -0.70
    Commission
    -0.66
     Participation
    -0.66
     Cannon
    -0.66
    mir
    -0.63
     timeline
    -0.62
     Thunder
    -0.62
     Reneg
    -0.62
     00
    -0.61
    planet
    -0.61
    POSITIVE LOGITS
    fts
    4.07
    kson
    1.83
    etts
    1.17
    fters
    1.11
    agog
    1.08
    iques
    1.07
     Oaks
    1.03
    ople
    1.00
    achusetts
    0.95
    culosis
    0.93
    Activations Density 0.034%

    No Known Activations