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. 11-RES-JB
    6. 659
    Prev
    Next
    INDEX
    Explanations

    instances where actions are being carried out with the help or assistance of others

    oai_token-act-pair · gpt-3.5-turbo

    phrases involving methods or procedures accompanied by the word "with."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    deen
    -0.74
    Reviewer
    -0.70
    gamer
    -0.69
    fuck
    -0.67
    MORE
    -0.65
    TG
    -0.65
     Awakens
    -0.62
     Watching
    -0.60
    camp
    -0.60
    film
    -0.60
    POSITIVE LOGITS
     regard
    1.36
     regards
    1.24
     respect
    1.21
    stood
    1.20
    drawn
    1.19
     impunity
    1.18
    draw
    1.11
    standing
    1.03
     assistance
    1.00
     attendant
    0.99
    Activations Density 0.164%

    No Known Activations