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. 10-RES-JB
    6. 15066
    Prev
    Next
    INDEX
    Explanations

    mentions of the word "Metal" with variations in capitalization and suffixes

    oai_token-act-pair · gpt-3.5-turbo

    references to "metal" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Suc
    -0.72
    zee
    -0.70
     Rowling
    -0.67
    erate
    -0.66
    ========
    -0.65
     Nurs
    -0.63
     Enlight
    -0.62
     Grande
    -0.62
    yip
    -0.62
    renheit
    -0.60
    POSITIVE LOGITS
    anguage
    1.67
     detectors
    1.35
     detector
    1.21
     oxide
    1.20
    works
    1.10
    working
    1.10
    heads
    1.06
    bending
    1.03
    worker
    0.99
    workers
    0.98
    Activations Density 0.018%

    No Known Activations