© 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. 8-RES-JB
    6. 7787
    Prev
    Next
    INDEX
    Explanations

    words relating to nighttime or late activities

    oai_token-act-pair · gpt-3.5-turbo

    references to late-night television shows and their hosts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    achu
    -0.85
     nomine
    -0.72
     intangible
    -0.71
     brim
    -0.70
     tolerance
    -0.69
     DOI
    -0.68
    âķIJ
    -0.67
     mirrors
    -0.64
     Rica
    -0.64
     gelatin
    -0.64
    POSITIVE LOGITS
    earth
    0.92
    angled
    0.91
    night
    0.90
    Earth
    0.85
    success
    0.84
    middle
    0.84
    period
    0.81
    early
    0.80
    Georg
    0.80
    model
    0.79
    Activations Density 0.066%

    No Known Activations