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. 3-RES-JB
    6. 22177
    Prev
    Next
    INDEX
    Explanations

    proper nouns or names starting with 'E' and followed by a lowercase letter 'l'

    oai_token-act-pair · gpt-3.5-turbo

    instances of the prefix "El" followed by various terms

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     RSS
    -0.71
     Flavoring
    -0.69
     cannabin
    -0.67
    mercial
    -0.64
     BW
    -0.62
     QR
    -0.62
    wcs
    -0.61
     ACS
    -0.61
    alter
    -0.61
    ongyang
    -0.61
    POSITIVE LOGITS
    otte
    0.88
    abeth
    0.79
    abet
    0.75
    quez
    0.75
    onian
    0.74
    irs
    0.72
    rette
    0.70
    abal
    0.70
    oche
    0.70
    otide
    0.68
    Activations Density 0.121%

    No Known Activations