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. 21556
    Prev
    Next
    INDEX
    Explanations

    the name "Harper" or references related to the person named Harper

    oai_token-act-pair · gpt-3.5-turbo

    mentions of individuals named Harper or Ambrose

    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
    CAST
    -0.69
     vent
    -0.68
    stration
    -0.65
    IJ
    -0.64
    Authent
    -0.62
     Sok
    -0.61
     lava
    -0.60
     liquid
    -0.60
     (=
    -0.59
     server
    -0.58
    POSITIVE LOGITS
     Harper
    4.02
     Mulcair
    1.85
     Trudeau
    1.76
    udeau
    1.62
    Har
    1.61
     Duffy
    1.35
     Abbott
    1.34
     Ottawa
    1.34
     Conservatives
    1.24
     Canadians
    1.24
    Activations Density 0.014%

    No Known Activations