© 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. 15032
    Prev
    Next
    INDEX
    Explanations

    terms related to persistence, existence, return, and consequence

    oai_token-act-pair · gpt-3.5-turbo

    references to dynamics or characteristics of existence and change

    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
    iHUD
    -0.75
    bara
    -0.72
     GER
    -0.65
    ebus
    -0.62
    erity
    -0.61
    ©¶æ
    -0.61
    ij士
    -0.61
    CCC
    -0.60
     Conan
    -0.60
     Garage
    -0.59
    POSITIVE LOGITS
     themselves
    1.30
     selves
    1.18
     individually
    1.11
    selves
    1.11
     counterparts
    0.80
     harmless
    0.77
     replacements
    0.73
    geries
    0.73
     unmarked
    0.72
     emit
    0.71
    Activations Density 0.544%

    No Known Activations