© 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. 6271
    Prev
    Next
    INDEX
    Explanations

    mathematical or geometric terms, specifically focusing on the spiral shape

    oai_token-act-pair · gpt-3.5-turbo

    references to images and visual elements, particularly in the context of a spiral motif

    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
    esy
    -0.97
    hew
    -0.91
    ety
    -0.91
    resent
    -0.88
    lehem
    -0.87
    ors
    -0.87
    aret
    -0.86
    earing
    -0.84
    olulu
    -0.84
    ellery
    -0.83
    POSITIVE LOGITS
    GROUP
    0.77
     Desk
    0.75
    BIL
    0.74
     Group
    0.71
    ature
    0.70
     Companies
    0.67
     Seas
    0.66
     Nano
    0.66
    acters
    0.65
     Ocean
    0.65
    Activations Density 0.036%

    No Known Activations