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. 1406
    Prev
    Next
    INDEX
    Explanations

    phrases related to large quantities or magnitudes

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "sheer" and its variations to denote magnitude or intensity

    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
    anwhile
    -0.83
    afort
    -0.75
    uberty
    -0.73
    azel
    -0.73
    itsch
    -0.69
    azine
    -0.68
    bye
    -0.68
    sem
    -0.67
    roe
    -0.66
    duct
    -0.66
    POSITIVE LOGITS
     quantity
    1.08
     volume
    1.06
     unbeliev
    1.02
     coincidence
    0.99
     numbers
    0.94
     amount
    0.93
     quantities
    0.91
     necessity
    0.90
     impossibility
    0.88
     sheer
    0.88
    Activations Density 0.023%

    No Known Activations