© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 20304
    Prev
    Next
    INDEX
    Explanations

    words related to comfort or physical well-being

    oai_token-act-pair · gpt-3.5-turbo

    keywords related to comfort and coziness

    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
    Not in Any Lists

    No Comments

    Negative Logits
     Kubrick
    -0.72
     regard
    -0.67
    uthor
    -0.66
     infringing
    -0.60
     inund
    -0.60
    uckland
    -0.59
     subclass
    -0.58
     gratification
    -0.58
     harbor
    -0.57
     commencement
    -0.57
    POSITIVE LOGITS
    ptroller
    1.36
    forts
    1.34
    ission
    1.13
    rade
    1.06
    meric
    1.06
    frey
    1.05
    etary
    1.02
    ittee
    1.00
    fy
    1.00
    fortable
    0.99
    Activations Density 0.015%

    No Known Activations