© 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
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 11-RES-JB
    6. 2069
    Prev
    Next
    INDEX
    Explanations

    instances of phrases declaring a certain sentiment or reaction

    oai_token-act-pair · gpt-3.5-turbo

    expressions of emotions and sentiments of gratitude, commitment, and concern

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     lasts
    -0.70
     Topics
    -0.66
    ãĤ¨ãĥ«
    -0.64
    acters
    -0.63
     Personality
    -0.62
    oras
    -0.62
    otle
    -0.62
    asma
    -0.60
    VERTISEMENT
    -0.59
    ologne
    -0.59
    POSITIVE LOGITS
     pleased
    1.36
     grateful
    1.34
     saddened
    1.29
     appalled
    1.26
     thankful
    1.26
     proud
    1.21
     thrilled
    1.20
     disappointed
    1.20
     delighted
    1.19
     confident
    1.17
    Activations Density 0.111%

    No Known Activations