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. 4-RES-JB
    6. 12795
    Prev
    Next
    INDEX
    Explanations

    sequences related to the Fibonacci sequence

    oai_token-act-pair · gpt-3.5-turbo

    sequences related to the Fibonacci sequence and associated terms

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

    No Comments

    Negative Logits
    oS
    -0.73
     Sov
    -0.71
    rez
    -0.70
    IPS
    -0.68
     tert
    -0.65
     Jag
    -0.65
    ontent
    -0.64
     STATS
    -0.64
     Berger
    -0.62
    fol
    -0.61
    POSITIVE LOGITS
     dolls
    0.81
    hler
    0.80
    mallow
    0.77
     padd
    0.77
     parenting
    0.74
    Parents
    0.74
    Redditor
    0.73
    Prin
    0.72
    uyomi
    0.71
     moms
    0.71
    Activations Density 0.150%

    No Known Activations