© 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. 0-RES-JB
    11. 15625
    Prev
    Next
    INDEX
    Explanations

    the word "drown" in various contexts

    oai_token-act-pair · gpt-3.5-turbo

    words and phrases related to drowning and water-related tragedies

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Whe
    -0.72
    iry
    -0.71
    roma
    -0.69
    GN
    -0.68
     cowork
    -0.68
    whe
    -0.65
     Lerner
    -0.64
    Whe
    -0.64
    Ang
    -0.64
     Romo
    -0.64
    POSITIVE LOGITS
     drown
    3.48
     drowning
    3.46
     drowned
    3.24
     submerged
    1.98
     submer
    1.84
     raft
    1.64
     flooded
    1.61
     inund
    1.58
     swim
    1.56
     swimming
    1.54
    Activations Density 0.036%

    No Known Activations