© 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. 10-RES-JB
    11. 15019
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "were" in sentences

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "were."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Piercing
    -0.74
    hma
    -0.73
    lag
    -0.71
    mare
    -0.65
     Solution
    -0.64
    coe
    -0.64
    Rating
    -0.64
     Label
    -0.63
    izable
    -0.63
    oire
    -0.62
    POSITIVE LOGITS
    wolves
    1.59
    wolf
    1.12
     subjected
    0.95
     evacuated
    0.94
     able
    0.93
    hes
    0.93
     unable
    0.93
     originally
    0.91
     initially
    0.91
     massac
    0.90
    Activations Density 0.257%

    No Known Activations