© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    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. 10650
    Prev
    Next
    INDEX
    Explanations

    phrases related to principles and values, possibly with a focus on discussions of public virtues

    oai_token-act-pair · gpt-3.5-turbo

    concepts related to instincts or basic human behaviors

    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
    itatively
    -0.69
    astical
    -0.69
    exit
    -0.67
    THER
    -0.67
     unspecified
    -0.66
    iliary
    -0.65
     unnamed
    -0.63
    Detailed
    -0.62
    mini
    -0.62
    antically
    -0.62
    POSITIVE LOGITS
     Thankfully
    0.82
     Often
    0.76
     Unfortunately
    0.74
     Hence
    0.73
    Often
    0.72
     Luckily
    0.72
     Accordingly
    0.72
    Thankfully
    0.72
     Especially
    0.71
     Therefore
    0.71
    Activations Density 0.969%

    No Known Activations