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. 4-RES-JB
    6. 23285
    Prev
    Next
    INDEX
    Explanations

    Nin-related terms, possibly referring to the word "Ninja."

    oai_token-act-pair · gpt-3.5-turbo

    references to "Nin" or "ninja" in various contexts

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

    No Comments

    Negative Logits
    inventoryQuantity
    -0.76
     ASC
    -0.68
     Miranda
    -0.67
     mileage
    -0.66
     COUNTY
    -0.66
     carriers
    -0.65
     HIP
    -0.64
     PRESS
    -0.63
     DRAG
    -0.63
     Hilton
    -0.62
    POSITIVE LOGITS
    eteen
    1.87
    eteenth
    1.67
    eties
    1.52
    ety
    1.41
    jas
    1.41
    jah
    1.14
    ovember
    1.10
    ja
    1.09
    ormal
    1.07
    omore
    1.01
    Activations Density 0.013%

    No Known Activations