© 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. 11-RES-JB
    11. 1901
    Prev
    Next
    INDEX
    Explanations

    instances of the word "pol" or variations like "police," "policy" or similar, potentially indicating a focus on terms related to politics or law enforcement

    oai_token-act-pair · gpt-3.5-turbo

    the word "police" and its variants in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    uring
    -0.70
     Revel
    -0.67
    rers
    -0.65
    INESS
    -0.65
     writ
    -0.61
     Druid
    -0.61
    EMA
    -0.57
     significant
    -0.55
     distraction
    -0.55
    ORGE
    -0.55
    POSITIVE LOGITS
    itely
    1.23
    arity
    1.16
    anski
    1.10
    ikarp
    1.07
    ishes
    1.04
    tical
    1.04
    ka
    1.03
    ishing
    1.02
    enta
    1.02
    ishers
    1.01
    Activations Density 0.031%

    No Known Activations