© 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. 3-RES-JB
    11. 23751
    Prev
    Next
    INDEX
    Explanations

    words related to security or protection

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "guard" and its variations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    BLE
    -0.81
    CAST
    -0.72
    KEN
    -0.70
    ãĥ£
    -0.70
    arenthood
    -0.65
    ANK
    -0.64
    theless
    -0.64
    à¤
    -0.64
    FORM
    -0.63
    ctive
    -0.62
    POSITIVE LOGITS
     guard
    0.98
     guarding
    0.93
     guards
    0.90
    guards
    0.86
    rail
    0.86
    ages
    0.85
     Guard
    0.83
     Guards
    0.81
     guarded
    0.80
    andowski
    0.80
    Activations Density 0.015%

    No Known Activations