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. 18475
    Prev
    Next
    INDEX
    Explanations

    references to domestic activities and daily routines

    oai_token-act-pair · gpt-3.5-turbo

    references to personal possessions or ownership

    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
    hots
    -0.79
    nir
    -0.78
    thood
    -0.74
    upload
    -0.71
    isan
    -0.71
     namely
    -0.71
     Ultimately
    -0.70
     allowing
    -0.69
    rame
    -0.69
    etheless
    -0.69
    POSITIVE LOGITS
     occasional
    1.25
     slightest
    1.23
     coolest
    1.15
     hottest
    1.12
     Kardash
    1.11
     nearest
    1.10
     proverbial
    1.07
     dreaded
    1.06
     smallest
    1.05
     fridge
    1.04
    Activations Density 0.355%

    No Known Activations