© 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. 8-RES-JB
    11. 2045
    Prev
    Next
    INDEX
    Explanations

    references to technological specifications related to 4G, LTE, and 4K content

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the number "4," especially in a context related to technology and products

    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
    esville
    -0.71
    vier
    -0.70
     Whitman
    -0.67
    UTION
    -0.65
    idence
    -0.64
    aries
    -0.61
    venue
    -0.60
    age
    -0.59
    becca
    -0.58
    iterator
    -0.58
    POSITIVE LOGITS
    teenth
    1.39
    teen
    1.35
    eva
    1.04
    some
    1.00
    Kids
    0.92
    Chan
    0.90
    square
    0.87
    chan
    0.86
    hyde
    0.85
    cyl
    0.83
    Activations Density 0.138%

    No Known Activations