© 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. 9-RES-JB
    11. 15452
    Prev
    Next
    INDEX
    Explanations

    contractions and negations

    oai_token-act-pair · gpt-3.5-turbo

    negative expressions of uncertainty or lack of understanding

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 9-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    9
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Flowers
    -0.65
     DAC
    -0.64
    agonal
    -0.64
    ocious
    -0.63
     Deer
    -0.63
     Ammo
    -0.63
     Sold
    -0.62
     Warm
    -0.61
    opl
    -0.59
     Coleman
    -0.58
    POSITIVE LOGITS
     emulate
    1.03
     deem
    0.99
     cherish
    0.82
     deems
    0.81
     despise
    0.76
    rina
    0.74
     exploit
    0.73
     termed
    0.73
     inhabit
    0.72
     sorely
    0.71
    Activations Density 0.309%

    No Known Activations