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. 3-RES-JB
    6. 2679
    Prev
    Next
    INDEX
    Explanations

    proper nouns or terms related to the word "Pres" and variations of it

    oai_token-act-pair · gpt-3.5-turbo

    repeated mentions of the term "Pres" or variations thereof, likely indicating references to a specific title or individual represented as "President."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ItemTracker
    -0.76
     Abyssal
    -0.75
    rawling
    -0.68
     Hebdo
    -0.67
    RANT
    -0.67
    rawl
    -0.65
    ISM
    -0.65
    ļéĨĴ
    -0.64
     Caribbean
    -0.61
     brackets
    -0.60
    POSITIVE LOGITS
    erves
    1.05
     Pres
    1.05
    cription
    1.05
    chool
    1.04
    Pres
    0.97
    ervation
    0.90
    pring
    0.89
    pres
    0.87
    mond
    0.87
    erving
    0.86
    Activations Density 0.008%

    No Known Activations