© 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. 16376
    Prev
    Next
    INDEX
    Explanations

    mentions or phrases related to activities or entities being run by a specific authority or organization

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "run", indicating references to running events or activities

    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
    hower
    -0.73
    ĵĺ
    -0.71
     Ink
    -0.68
    ¿½
    -0.68
    EStream
    -0.67
    ortium
    -0.67
    appy
    -0.65
     Brist
    -0.65
    ++++
    -0.65
     Hots
    -0.65
    POSITIVE LOGITS
    ners
    1.20
    escape
    1.14
    swick
    0.98
    ctors
    0.96
    nings
    0.95
    nels
    0.93
    estones
    0.88
    runners
    0.88
    nell
    0.82
    ters
    0.82
    Activations Density 0.019%

    No Known Activations