© 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. 7233
    Prev
    Next
    INDEX
    Explanations

    mentions of age and familial relationships

    oai_token-act-pair · gpt-3.5-turbo

    numerical references to the age or quantity of individuals or animals

    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
     guiName
    -0.74
     Tradable
    -0.68
    igham
    -0.67
     Indust
    -0.64
    cair
    -0.62
    ording
    -0.60
     itself
    -0.60
    vell
    -0.60
    rity
    -0.59
    atem
    -0.59
    POSITIVE LOGITS
     counterparts
    0.99
     predecessors
    0.97
     namesake
    0.91
    favorite
    0.91
     daughters
    0.89
     favorite
    0.88
     own
    0.86
     birthday
    0.85
     cents
    0.84
     foray
    0.84
    Activations Density 0.115%

    No Known Activations