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. 0-RES-JB
    6. 22888
    Prev
    Next
    INDEX
    Explanations

    computer coding terms and commands, specifically those related to file handling and coding instructions

    oai_token-act-pair · gpt-3.5-turbo

    underscore characters in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Ninth
    -0.66
     heck
    -0.65
    reau
    -0.61
     rounding
    -0.58
     famed
    -0.58
     Sands
    -0.54
     numerous
    -0.53
     nuns
    -0.53
     unaccount
    -0.52
     "â̦
    -0.52
    POSITIVE LOGITS
    _
    3.18
    _(
    1.91
    _-_
    1.68
    __
    1.65
    _{
    1.58
    _-
    1.57
    ___
    1.53
    _.
    1.53
    _>
    1.39
    ._
    1.32
    Activations Density 0.035%

    No Known Activations