© 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. 0-RES-JB
    11. 10430
    Prev
    Next
    INDEX
    Explanations

    the word "Instead"

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "Instead" indicating a contrast or shift in perspective

    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
     unknown
    -0.66
     fluct
    -0.65
     dating
    -0.65
     comm
    -0.60
     receiving
    -0.60
     asc
    -0.59
     various
    -0.58
     varying
    -0.58
     margin
    -0.58
     mur
    -0.57
    POSITIVE LOGITS
     Instead
    3.55
    Instead
    3.13
     Rather
    2.38
    instead
    2.26
    Rather
    2.09
     instead
    2.09
    rather
    1.58
     Nope
    1.50
     Thankfully
    1.38
     Nonetheless
    1.38
    Activations Density 0.018%

    No Known Activations