© 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
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 4-RES-JB
    6. 15234
    Prev
    Next
    INDEX
    Explanations

    positive and luxurious descriptive words

    oai_token-act-pair · gpt-3.5-turbo

    words and phrases associated with luxury and opulence

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

    No Comments

    Negative Logits
    utor
    -0.72
    wik
    -0.63
    TW
    -0.63
    itans
    -0.63
    reon
    -0.62
     Brend
    -0.60
     unilaterally
    -0.60
     administrator
    -0.60
    abus
    -0.60
     incorrectly
    -0.59
    POSITIVE LOGITS
     splend
    0.84
     elegance
    0.82
    ocl
    0.81
     perfume
    0.76
     dazz
    0.75
    puff
    0.74
     beauty
    0.73
    ishly
    0.73
     boutique
    0.72
     fragrance
    0.71
    Activations Density 0.098%

    No Known Activations