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. 8-RES-JB
    6. 19751
    Prev
    Next
    INDEX
    Explanations

    words related to the word "cyclone"

    oai_token-act-pair · gpt-3.5-turbo

    terms related to cycling and bicycles

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    FORE
    -0.85
    IELD
    -0.77
     Cheong
    -0.67
    GROUND
    -0.65
     Bris
    -0.64
    fitting
    -0.63
    AMI
    -0.62
     Malik
    -0.62
    OURCE
    -0.61
    LESS
    -0.61
    POSITIVE LOGITS
    ical
    1.15
    umn
    1.09
    omatic
    1.07
    onel
    0.98
    oton
    0.97
    ones
    0.97
    icism
    0.94
    ozo
    0.94
    ophon
    0.93
    onic
    0.93
    Activations Density 0.035%

    No Known Activations