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. 905
    Prev
    Next
    INDEX
    Explanations

    mentions of grandfathers

    oai_token-act-pair · gpt-3.5-turbo

    references to familial relationships, specifically focusing on grandfathers and grandmothers

    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
    IL
    -0.68
    scape
    -0.67
    cock
    -0.66
    waves
    -0.66
     Parsons
    -0.65
    eth
    -0.63
     Scene
    -0.62
    ily
    -0.62
     Slay
    -0.62
     CP
    -0.61
    POSITIVE LOGITS
     grandfather
    3.24
     grandparents
    3.16
     grandmother
    3.15
     grandma
    2.54
     grandchildren
    1.97
     grandson
    1.94
     aunt
    1.93
     granddaughter
    1.92
     uncle
    1.70
    grand
    1.61
    Activations Density 0.021%

    No Known Activations