© 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. 10338
    Prev
    Next
    INDEX
    Explanations

    words related to biology, evolution, or transformation

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of specific morphological forms of the word "trope" and related concepts

    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
    Legend
    -0.81
    ruck
    -0.80
     Halls
    -0.76
     Chaff
    -0.74
    ascar
    -0.74
     Haunted
    -0.74
     Twain
    -0.73
    Church
    -0.72
     Fres
    -0.71
     Fitz
    -0.71
    POSITIVE LOGITS
     tro
    1.78
     cellular
    1.56
     inflammation
    1.37
     candid
    1.30
     mitochondrial
    1.28
    llular
    1.27
     inflammatory
    1.26
     Cellular
    1.15
     PL
    1.14
     vitro
    1.14
    Activations Density 0.044%

    No Known Activations