© 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
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 0-RES-JB
    6. 16660
    Prev
    Next
    INDEX
    Explanations

    phrases indicating immediacy or quick action

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "instantly" and other adverbs indicating immediacy

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     caution
    -0.67
    osures
    -0.64
     Bottom
    -0.62
     contingency
    -0.61
    spection
    -0.61
     caveat
    -0.60
    osa
    -0.59
    ays
    -0.58
     CFR
    -0.58
    vere
    -0.57
    POSITIVE LOGITS
     instantly
    3.49
     immediately
    2.04
     instant
    1.76
     effortlessly
    1.74
     automatically
    1.66
     quickly
    1.62
     instinctively
    1.59
     swiftly
    1.57
     instantaneous
    1.53
     promptly
    1.51
    Activations Density 0.012%

    No Known Activations