© 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. 10-RES-JB
    11. 809
    Prev
    Next
    INDEX
    Explanations

    positive sentiment words expressing strong emotions, particularly happiness and excitement

    oai_token-act-pair · gpt-3.5-turbo

    emotions expressed in the context of happiness or positive sentiments

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    •layer 10 - joy

    No Comments

    Negative Logits
     disappearing
    -0.64
     vanish
    -0.62
     Stealth
    -0.61
     extinction
    -0.60
    amins
    -0.59
    Afee
    -0.58
     Fail
    -0.57
     disappear
    -0.57
    geries
    -0.56
     killer
    -0.56
    POSITIVE LOGITS
     thankful
    1.62
     grateful
    1.61
     pleased
    1.59
     glad
    1.50
     proud
    1.43
     excited
    1.41
     saddened
    1.37
     disappointed
    1.35
     impressed
    1.33
     happy
    1.30
    Activations Density 0.089%

    No Known Activations