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. 21482
    Prev
    Next
    INDEX
    Explanations

    phrases containing the name "Ert" followed by a number, potentially related to sports players or other individuals

    oai_token-act-pair · gpt-3.5-turbo

    references to a specific individual named "Gert" or variations of that name

    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
    ntax
    -0.70
    daq
    -0.68
    AW
    -0.67
    Story
    -0.65
    CLOSE
    -0.64
    IDS
    -0.64
    hyde
    -0.63
    Ĥª
    -0.62
    ĪĴ
    -0.61
    TOR
    -0.61
    POSITIVE LOGITS
    ificate
    1.07
    urbed
    1.03
    itude
    0.98
    ainer
    0.98
    ained
    0.96
    aining
    0.95
    ieth
    0.95
    ruck
    0.94
    ogether
    0.93
    rait
    0.92
    Activations Density 0.048%

    No Known Activations