© 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. 4-RES-JB
    11. 20185
    Prev
    Next
    INDEX
    Explanations

    phrases with the word "said" indicating a quote or statement

    oai_token-act-pair · gpt-3.5-turbo

    instances of reported speech or quotations

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

    No Comments

    Negative Logits
    otion
    -0.74
    EMBER
    -0.64
    pleting
    -0.64
    xtap
    -0.63
    çIJ
    -0.62
    \/\/
    -0.62
    physical
    -0.61
    à¦
    -0.61
    arettes
    -0.60
    izont
    -0.60
    POSITIVE LOGITS
     goodbye
    0.69
     bluntly
    0.67
     yesterday
    0.65
    doms
    0.61
     earlier
    0.60
     emphatically
    0.59
     sarcast
    0.58
     aloud
    0.58
     prof
    0.56
     repeatedly
    0.55
    Activations Density 0.191%

    No Known Activations