Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 9276
    Prev
    Next
    INDEX
    Explanations

    phrases that begin with "Upon"

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "Upon" and its variants

    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
    emic
    -0.76
     outl
    -0.65
    rast
    -0.63
     marginal
    -0.63
     eclips
    -0.63
     intens
    -0.62
     bigger
    -0.62
     lobbyists
    -0.61
     minority
    -0.60
     fres
    -0.59
    POSITIVE LOGITS
    Upon
    3.38
     Upon
    2.85
    upon
    2.53
     upon
    1.39
     Immediately
    1.17
     Afterwards
    1.15
     Whilst
    1.04
     Accordingly
    0.93
    soever
    0.90
    Within
    0.86
    Activations Density 0.024%

    No Known Activations