© 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. 3-RES-JB
    11. 12353
    Prev
    Next
    INDEX
    Explanations

    mentions of someone being officially declared deceased or having died

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "pronounced" related to death or declaration

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    committee
    -0.78
    ionage
    -0.78
    hatt
    -0.71
    chedel
    -0.70
    allow
    -0.68
    sol
    -0.67
    adr
    -0.66
    Jackson
    -0.66
     Camer
    -0.65
    chel
    -0.65
    POSITIVE LOGITS
     pronounced
    0.98
     pronounce
    0.93
     pronunciation
    0.85
    aneously
    0.80
     audible
    0.76
     conson
    0.75
     spelling
    0.75
    senal
    0.73
     vowel
    0.73
    Þ
    0.72
    Activations Density 0.009%

    No Known Activations