© 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
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 9-RES-JB
    11. 3212
    Prev
    Next
    INDEX
    Explanations

    the word "key."

    oai_token-act-pair · gpt-3.5-turbo

    words that indicate essential elements or critical insights in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 9-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    9
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Leone
    -0.73
    ©¶æ
    -0.70
     Tsukuyomi
    -0.69
    uum
    -0.68
    aving
    -0.67
     Kitt
    -0.66
     Mile
    -0.65
     Claud
    -0.63
     Canaver
    -0.63
     Zur
    -0.62
    POSITIVE LOGITS
    stone
    1.49
    stroke
    1.21
    stones
    1.20
    stro
    1.12
    hole
    1.10
    notes
    1.09
    */(
    1.05
    binding
    1.03
    chain
    1.01
     ingredient
    0.99
    Activations Density 0.030%

    No Known Activations