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. 21617
    Prev
    Next
    INDEX
    Explanations

    mentions of specific time or location details in narratives

    oai_token-act-pair · gpt-3.5-turbo

    activities related to shopping and family

    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
     legions
    -0.77
     cures
    -0.69
     definitive
    -0.66
     plaque
    -0.66
     captcha
    -0.66
     vow
    -0.66
    DOS
    -0.65
     renders
    -0.64
     genomes
    -0.64
     disse
    -0.64
    POSITIVE LOGITS
    livious
    0.84
     partying
    0.77
     picnic
    0.77
     shopping
    0.76
     asleep
    0.75
    ercise
    0.73
    chard
    0.72
     downstairs
    0.72
    girlfriend
    0.71
     robbery
    0.71
    Activations Density 0.538%

    No Known Activations