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. 19946
    Prev
    Next
    INDEX
    Explanations

    references to a specific person's name, "Benedict," which appears in various contexts

    oai_token-act-pair · gpt-3.5-turbo

    mentions of Pope Benedict and associated topics

    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
    perse
    -0.75
    EAR
    -0.75
    claimer
    -0.73
    ODUCT
    -0.71
    MAG
    -0.68
    porting
    -0.67
    markets
    -0.66
    place
    -0.65
    NING
    -0.65
    nesota
    -0.64
    POSITIVE LOGITS
     Cumber
    1.29
     XVI
    1.21
     Benedict
    1.14
    itial
    0.93
    olini
    0.88
    itude
    0.86
    icum
    0.85
    inas
    0.84
     Arnold
    0.81
    zman
    0.80
    Activations Density 0.004%

    No Known Activations