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. 214
    Prev
    Next
    INDEX
    Explanations

    mentions of the term "Galaxy", likely referring to the technology brand or possibly the term in a science-fiction context

    oai_token-act-pair · gpt-3.5-turbo

    mentions of the term "Galaxy" 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
     Alison
    -0.69
     trad
    -0.66
    mit
    -0.66
    houn
    -0.65
    rig
    -0.65
     rehabilit
    -0.64
     pow
    -0.64
    bral
    -0.63
    fet
    -0.63
     repro
    -0.62
    POSITIVE LOGITS
     Galaxy
    4.15
     galaxy
    1.99
     Cosmos
    1.65
    axy
    1.59
     Milky
    1.58
     Droid
    1.54
    Gal
    1.52
     Samsung
    1.52
    Samsung
    1.50
     Galactic
    1.48
    Activations Density 0.022%

    No Known Activations