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. 10-RES-JB
    6. 101
    Prev
    Next
    INDEX
    Explanations

    JavaScript code snippets commonly used to interact with a user database

    oai_token-act-pair · gpt-3.5-turbo

    references to user management and operations in programming

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    •GPT2 L10 resid: 'LRRRRRLLRRRL'[-1]
    •GPT2 L10 resid: 'LRRRRRLLRRRL'[:-2]
    •GPT2 L10 resid: 'LRRRRRLLRRRL'[:-3]

    No Comments

    Negative Logits
    taboola
    -0.73
     apologies
    -0.70
    theless
    -0.65
     jaws
    -0.64
    anwhile
    -0.64
     lawy
    -0.64
     mosqu
    -0.63
     rodents
    -0.62
    yond
    -0.62
     trolls
    -0.62
    POSITIVE LOGITS
    ()
    1.26
    Property
    1.25
    Async
    1.24
    ();
    1.23
    ById
    1.23
    Object
    1.21
    ())
    1.20
    String
    1.19
    Parameters
    1.17
    Changed
    1.15
    Activations Density 0.054%

    No Known Activations