© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 10-RES-JB
    11. 15987
    Prev
    Next
    INDEX
    Explanations

    phrases with the pattern "to-X-to", where X can be a word or empty

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to vehicle-related topics or discussions

    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
     flares
    -0.69
     Hung
    -0.65
     modulation
    -0.64
     dependent
    -0.63
     affirmation
    -0.63
     messing
    -0.63
     sooner
    -0.63
     exploits
    -0.63
     Wiley
    -0.63
     indications
    -0.62
    POSITIVE LOGITS
    date
    1.19
    prison
    1.17
    advertising
    1.17
    itiveness
    1.15
    market
    1.13
    life
    1.12
    toe
    1.11
    gas
    1.10
    be
    1.10
    speech
    1.10
    Activations Density 0.025%

    No Known Activations