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. 17429
    Prev
    Next
    INDEX
    Explanations

    words related to radiation and radioactive materials

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "rad" and its variations in different 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
     Faul
    -0.66
     Emacs
    -0.66
     bash
    -0.64
     Baton
    -0.64
     Fifa
    -0.63
     gorilla
    -0.62
     Haiti
    -0.62
     finish
    -0.62
     Firefox
    -0.61
     butterfly
    -0.61
    POSITIVE LOGITS
    rad
    4.73
    Rad
    1.90
    rade
    1.54
    rat
    1.42
     rad
    1.40
    ras
    1.31
    rador
    1.26
    rogens
    1.24
    rod
    1.21
     Rad
    1.20
    Activations Density 0.016%

    No Known Activations