Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 3003
    Prev
    Next
    INDEX
    Explanations

    words related to technology, services, and locations

    oai_token-act-pair · gpt-3.5-turbo

    nouns or terms that refer to specific roles or professions

    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
    •single token firings
    Individually fires on capitlized " Guests", "Players", " Patients", " Players", " Drivers", " Owners", "Companies", "Engineers", ...

    No Comments

    Negative Logits
    ãĥª
    -0.74
    SpaceEngineers
    -0.73
    Accessory
    -0.68
    SPONSORED
    -0.65
    Bey
    -0.61
    UCT
    -0.60
    ãĤ¦ãĤ¹
    -0.60
    Ïī
    -0.60
    ¾
    -0.58
    shapeshifter
    -0.57
    POSITIVE LOGITS
    hip
    1.04
    hips
    0.99
    wana
    0.98
    cius
    0.94
    cript
    0.93
    ource
    0.93
    lash
    0.90
     Called
    0.90
     Picks
    0.89
    afety
    0.87
    Activations Density 0.278%

    No Known Activations