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. 3453
    Prev
    Next
    INDEX
    Explanations

    proper nouns or names within a sentence

    oai_token-act-pair · gpt-3.5-turbo

    instances of parentheses

    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
     retard
    -0.84
     clin
    -0.84
     increment
    -0.82
     administration
    -0.72
     diving
    -0.71
     uncons
    -0.71
     accrued
    -0.71
     electr
    -0.70
     bunker
    -0.70
     tear
    -0.70
    POSITIVE LOGITS
    sic
    1.80
    formerly
    1.60
    which
    1.49
    aka
    1.48
    pron
    1.47
    pictured
    1.47
    meaning
    1.39
    see
    1.38
    later
    1.37
    whose
    1.37
    Activations Density 0.146%

    No Known Activations