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. 7911
    Prev
    Next
    INDEX
    Explanations

    words related to giving, providing, and allowing in the context of actions and policies

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to the act of giving or providing aid or resources

    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
    uden
    -0.88
    DragonMagazine
    -0.79
    æĪ¦
    -0.73
    76561
    -0.72
     Symbol
    -0.72
    uin
    -0.69
    walker
    -0.67
    UME
    -0.67
    Ãį
    -0.66
    ItemTracker
    -0.66
    POSITIVE LOGITS
     people
    1.08
     everyone
    1.06
     households
    1.06
     poorer
    1.06
     wealthier
    1.06
     consumers
    1.05
     kids
    1.04
     workplaces
    1.04
     individuals
    1.03
     employers
    1.03
    Activations Density 0.251%

    No Known Activations