© 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. 20244
    Prev
    Next
    INDEX
    Explanations

    phrases related to personal or organizational mottos

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to guiding principles or beliefs

    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
    NetMessage
    -0.89
    idan
    -0.78
    late
    -0.75
    ammy
    -0.74
    ockets
    -0.73
    akov
    -0.72
    aks
    -0.72
    lings
    -0.71
    adder
    -0.70
     Flavoring
    -0.69
    POSITIVE LOGITS
     motto
    1.08
     mantra
    1.02
     goodbye
    0.81
     maxim
    0.76
     slogan
    0.76
     dogma
    0.71
     ethos
    0.71
     mant
    0.71
     creed
    0.70
     refrain
    0.69
    Activations Density 0.018%

    No Known Activations