© 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. 3-RES-JB
    11. 15777
    Prev
    Next
    INDEX
    Explanations

    descriptions related to compactness

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "compact" and related terms

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Downloadha
    -0.83
    CVE
    -0.77
    opoulos
    -0.73
    olkien
    -0.67
    ãĤ´ãĥ³
    -0.67
    affer
    -0.67
    hler
    -0.66
    uthor
    -0.66
    cher
    -0.66
    hod
    -0.64
    POSITIVE LOGITS
     discs
    0.90
     compact
    0.89
     Compact
    0.85
    nesses
    0.81
    ness
    0.80
     footprint
    0.79
     fluorescent
    0.76
    ened
    0.74
     disks
    0.74
    arily
    0.73
    Activations Density 0.007%

    No Known Activations