© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    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. 2521
    Prev
    Next
    INDEX
    Explanations

    items or objects that are important or crucial to a person's identity or daily life

    oai_token-act-pair · gpt-3.5-turbo

    objects, documents, and concepts related to legality and personal property

    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
     oneself
    -0.73
    arnaev
    -0.71
    yrinth
    -0.66
     Emer
    -0.66
     yourselves
    -0.64
    htaking
    -0.64
    heric
    -0.63
    heny
    -0.63
     Canaver
    -0.63
     Hogan
    -0.62
    POSITIVE LOGITS
    mates
    1.09
     mates
    1.00
    mate
    1.00
     buddies
    0.83
     counterparts
    0.77
     credentials
    0.77
     colleague
    0.76
     holdings
    0.75
     woes
    0.75
     stash
    0.74
    Activations Density 0.427%

    No Known Activations