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. 22939
    Prev
    Next
    INDEX
    Explanations

    words related to defending or advocating for someone or something

    oai_token-act-pair · gpt-3.5-turbo

    references to defenders in various contexts

    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
    Jew
    -0.77
     Scale
    -0.70
    ancer
    -0.70
     sport
    -0.66
     Lifetime
    -0.63
     seat
    -0.63
    ibo
    -0.62
    mini
    -0.61
    inger
    -0.61
     Series
    -0.61
    POSITIVE LOGITS
     defenders
    3.67
     defenses
    2.07
     defences
    1.92
     attackers
    1.88
     opponents
    1.86
     detractors
    1.83
     blockers
    1.80
     Defenders
    1.71
     proponents
    1.62
     critics
    1.56
    Activations Density 0.024%

    No Known Activations