© 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. 3-RES-JB
    11. 5449
    Prev
    Next
    INDEX
    Explanations

    words related to expressing dissatisfaction or discontent

    oai_token-act-pair · gpt-3.5-turbo

    words associated with expressions of discomfort or emotional upheaval

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ŀ
    -0.73
     multipl
    -0.68
     redistributed
    -0.67
    ãĥ¼ãĥ«
    -0.67
    ãĤ©
    -0.66
     rgb
    -0.65
     kits
    -0.65
     levers
    -0.64
     geo
    -0.62
     smoot
    -0.62
    POSITIVE LOGITS
    ingly
    1.05
     loudly
    0.93
     aloud
    0.90
     angrily
    0.82
    selves
    0.80
    Against
    0.79
    antly
    0.78
    Continue
    0.77
    naires
    0.77
    ieu
    0.76
    Activations Density 0.163%

    No Known Activations