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. 24007
    Prev
    Next
    INDEX
    Explanations

    mentions indicating responsibility or action

    oai_token-act-pair · gpt-3.5-turbo

    instances of empty or missing content

    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
     millenn
    -1.19
     Instr
    -0.98
    govtrack
    -0.97
     enthusi
    -0.92
    iosyn
    -0.86
    aditional
    -0.79
     comr
    -0.78
    ikuman
    -0.77
    DragonMagazine
    -0.76
    luaj
    -0.76
    POSITIVE LOGITS
    <|endoftext|>
    1.06
    ignty
    0.91
    pmwiki
    0.84
    ð
    0.84
    etsk
    0.84
    ©¶æ
    0.82
    rawdownloadcloneembedreportprint
    0.81
    ebook
    0.79
    wcsstore
    0.78
    namese
    0.77
    Activations Density 0.782%

    No Known Activations