Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 5626
    Prev
    Next
    INDEX
    Explanations

    scientific terms related to measurements and calculations

    oai_token-act-pair · gpt-3.5-turbo

    terms related to various fields of measurement and analysis, especially those ending in "metric" or "ometry"

    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
     Savannah
    -0.65
    UU
    -0.61
     spirit
    -0.61
     Welch
    -0.61
     Verb
    -0.60
     resort
    -0.59
     bye
    -0.59
     conv
    -0.59
     Arabia
    -0.58
     resorts
    -0.57
    POSITIVE LOGITS
    ometric
    4.66
    ometry
    3.57
    omet
    2.69
    ometer
    2.06
    ometers
    1.95
    otropic
    1.40
    omorphic
    1.39
    ograms
    1.32
    ographic
    1.29
    okin
    1.26
    Activations Density 0.013%

    No Known Activations