© 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. 17267
    Prev
    Next
    INDEX
    Explanations

    the word "Hum" followed by a number, potentially related to the vehicle brand "Hummer."

    oai_token-act-pair · gpt-3.5-turbo

    references to human experiences and emotions

    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
    hips
    -0.78
    cape
    -0.75
     Dele
    -0.66
     Borough
    -0.64
     Leviathan
    -0.64
     DARK
    -0.63
     Peaks
    -0.61
     flagged
    -0.61
     Sacrament
    -0.59
     Trader
    -0.59
    POSITIVE LOGITS
    pty
    1.35
    orously
    1.22
    ility
    1.15
    ankind
    1.10
    iliation
    1.10
    mers
    1.09
    ulus
    1.07
    mus
    1.07
    ming
    1.05
    iliated
    1.04
    Activations Density 0.027%

    No Known Activations