© 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. 4-RES-JB
    11. 22593
    Prev
    Next
    INDEX
    Explanations

    phrases related to running out or depletion

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to running out of resources or supplies

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

    No Comments

    Negative Logits
    shaw
    -0.80
    ellation
    -0.71
    ettlement
    -0.69
    anooga
    -0.67
    gemony
    -0.67
    hetti
    -0.64
    antine
    -0.61
    rador
    -0.61
    ery
    -0.61
    antry
    -0.60
    POSITIVE LOGITS
    doors
    0.85
    fitted
    0.74
    casts
    0.66
    wards
    0.64
    reaching
    0.62
    OTAL
    0.61
    stretched
    0.61
    keeper
    0.61
    lier
    0.61
    flow
    0.60
    Activations Density 0.037%

    No Known Activations