Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APISteerSAE EvalsBlog/PodcastNEWSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlog/PodcastGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 6-RES-JB
    6. 17968
    Prev
    Next
    INDEX
    Explanations

    calls to action in marketing contexts

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to signing up for newsletters or subscriptions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    "up for", "up to", "in with" (after the word "sign"), in the context of commands to sign up or in; the phrase "Fight Back!"

    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 6-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    6
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    banks
    -0.79
    worth
    -0.77
    gery
    -0.76
    gow
    -0.74
    ļé
    -0.73
    geries
    -0.71
    bank
    -0.70
    dfx
    -0.69
    pher
    -0.68
    ../
    -0.67
    POSITIVE LOGITS
     Submit
    0.81
     SPACE
    0.69
     âĸ
    0.69
     Email
    0.69
     Aviation
    0.68
     Daily
    0.67
     alerts
    0.64
     LIKE
    0.63
     newsletter
    0.61
    MSN
    0.61
    Activations Density 0.087%

    No Known Activations