© 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. 8-RES-JB
    11. 17699
    Prev
    Next
    INDEX
    Explanations

    phrases indicating thoughts or beliefs

    oai_token-act-pair · gpt-3.5-turbo

    phrases that express thoughts, beliefs, or assumptions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    verting
    -0.77
    Rated
    -0.68
    hs
    -0.67
    md
    -0.67
    versions
    -0.65
    recorded
    -0.65
    unker
    -0.64
    draft
    -0.62
    jab
    -0.62
    ping
    -0.61
    POSITIVE LOGITS
    enance
    0.82
     logically
    0.79
     forgiven
    0.71
    DragonMagazine
    0.70
    gat
    0.68
     Esports
    0.66
     Santos
    0.65
     otherwise
    0.60
     Moreno
    0.60
     THAT
    0.59
    Activations Density 0.083%

    No Known Activations