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. 1580
    Prev
    Next
    INDEX
    Explanations

    abbreviated terms with periods as separators and potentially related keywords like 'reporter' or 'news' alongside

    oai_token-act-pair · gpt-3.5-turbo

    references to brands, companies, or specific cultural identifiers

    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
    catentry
    -0.92
    ilial
    -0.85
    ascript
    -0.79
     Faw
    -0.68
    ources
    -0.65
    âĸ¬
    -0.64
     Malf
    -0.64
    Versions
    -0.61
     Qiao
    -0.61
     inflamm
    -0.61
    POSITIVE LOGITS
     Lumpur
    1.08
    EStreamFrame
    0.93
    bye
    0.92
    EStream
    0.88
    enei
    0.85
    istan
    0.85
    ikuman
    0.81
    utra
    0.79
    nik
    0.75
    zie
    0.74
    Activations Density 0.355%

    No Known Activations