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. 4-RES-JB
    6. 301
    Prev
    Next
    INDEX
    Explanations

    words related to ethical, legal, and criminal matters

    oai_token-act-pair · gpt-3.5-turbo

    references to different forms of social and legal issues

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

    No Comments

    Negative Logits
     parentheses
    -0.62
     mur
    -0.60
    gra
    -0.59
     beh
    -0.59
    hung
    -0.58
    ays
    -0.57
    aug
    -0.56
    cknowled
    -0.55
     obser
    -0.55
     depress
    -0.54
    POSITIVE LOGITS
     whilst
    0.77
    aeda
    0.70
    Community
    0.70
     wherever
    0.69
    Eva
    0.68
    ãĢĤ
    0.67
    ocobo
    0.66
     cheaply
    0.65
     abroad
    0.65
    âĸĪâĸĪâĸĪâĸĪâĸĪâĸĪâĸĪâĸĪ
    0.64
    Activations Density 0.888%

    No Known Activations