© 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. 0-RES-JB
    11. 8231
    Prev
    Next
    INDEX
    Explanations

    instances of the word "warm."

    oai_token-act-pair · gpt-3.5-turbo

    references to warmth or warm temperatures

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ulo
    -0.73
    eda
    -0.72
    Effect
    -0.71
    uler
    -0.70
    RC
    -0.70
    ption
    -0.69
     MGM
    -0.65
    advertising
    -0.65
    laws
    -0.65
    Dou
    -0.65
    POSITIVE LOGITS
     warm
    3.67
    warm
    2.70
     Warm
    2.41
     warmer
    2.40
     warmed
    2.33
     warmth
    2.26
     warming
    1.86
     colder
    1.76
     chilly
    1.74
     cold
    1.72
    Activations Density 0.013%

    No Known Activations