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. 1-RES-JB
    6. 6229
    Prev
    Next
    INDEX
    Explanations

    personal pronouns or determiners indicating possession or relevance

    oai_token-act-pair · gpt-3.5-turbo

    the word "those" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    forth
    -0.64
     Ness
    -0.61
     Ridge
    -0.60
    ister
    -0.59
     Plus
    -0.59
    shapeshifter
    -0.59
    ocity
    -0.59
     Minion
    -0.59
     ..."
    -0.59
    Shock
    -0.59
    POSITIVE LOGITS
     pesky
    0.79
    etheless
    0.78
     kinds
    0.78
     sorts
    0.73
     resear
    0.70
    ratulations
    0.69
    types
    0.69
     interested
    0.68
     fellows
    0.68
     extrad
    0.68
    Activations Density 0.039%

    No Known Activations