© 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. 19124
    Prev
    Next
    INDEX
    Explanations

    references to the name "Celest"

    oai_token-act-pair · gpt-3.5-turbo

    mentions of the name "Cel" and its variations, likely referring to specific characters or entities related to the name

    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
     bee
    -0.68
     Bee
    -0.67
    ought
    -0.66
    soDeliveryDate
    -0.65
    sky
    -0.63
    intage
    -0.63
    reys
    -0.62
    kered
    -0.62
     resh
    -0.60
    atters
    -0.60
    POSITIVE LOGITS
     Cel
    3.96
     cel
    2.19
     Celt
    1.52
     Celest
    1.28
     Celtic
    1.22
     Cind
    1.21
     Kel
    1.14
     Ces
    1.04
     Carm
    1.02
     Belg
    1.01
    Activations Density 0.014%

    No Known Activations