© 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. 4-RES-JB
    11. 12002
    Prev
    Next
    INDEX
    Explanations

    acronyms starting with "HM" and a numerical identifier

    oai_token-act-pair · gpt-3.5-turbo

    references to governmental or royal entities, particularly those associated with "HM."

    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
     Vaugh
    -0.73
    ctors
    -0.73
     Races
    -0.72
     Franch
    -0.71
    aneously
    -0.70
    OPLE
    -0.70
     Ferr
    -0.67
     Haas
    -0.67
    eers
    -0.66
     slic
    -0.66
    POSITIVE LOGITS
    sonian
    0.83
    brance
    0.77
    doms
    0.77
    abyte
    0.75
    shire
    0.75
    imet
    0.74
    onder
    0.74
    conservancy
    0.74
    awei
    0.74
     HM
    0.72
    Activations Density 0.014%

    No Known Activations