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. 19909
    Prev
    Next
    INDEX
    Explanations

    content related to courses, lessons, or educational materials

    oai_token-act-pair · gpt-3.5-turbo

    instances of the phrase "of course."

    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
     reperto
    -0.71
    ocene
    -0.60
     ashore
    -0.58
    opus
    -0.58
     behave
    -0.57
    .","
    -0.57
     misunder
    -0.56
     violates
    -0.56
     slump
    -0.55
     entit
    -0.55
    POSITIVE LOGITS
     course
    1.69
    course
    1.45
     COUR
    1.23
     Course
    1.14
    sted
    1.01
    icial
    0.99
     note
    0.96
    elia
    0.95
    ortunately
    0.95
     paramount
    0.91
    Activations Density 0.028%

    No Known Activations