© 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. 12357
    Prev
    Next
    INDEX
    Explanations

    commands or requests to stop or terminate something

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "cease" and its variations in the context of stopping actions or events

    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
    rium
    -0.82
    ramid
    -0.81
    arse
    -0.74
    rics
    -0.74
    assium
    -0.74
    é¾įå
    -0.72
    ested
    -0.72
    odes
    -0.71
    OTE
    -0.71
    uries
    -0.71
    POSITIVE LOGITS
     cease
    0.86
     seiz
    0.83
    rences
    0.78
    cedes
    0.76
     hostilities
    0.73
     ceases
    0.73
    plin
    0.71
    ĸļ
    0.71
     abruptly
    0.71
    fires
    0.70
    Activations Density 0.009%

    No Known Activations