© 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. 10-RES-JB
    11. 2975
    Prev
    Next
    INDEX
    Explanations

    terms related to technology, particularly in the context of online platforms and communication

    oai_token-act-pair · gpt-3.5-turbo

    phrases involving connections or relationships between elements or concepts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     prepar
    -0.83
     skelet
    -0.83
     mechanically
    -0.80
     referen
    -0.79
     accomp
    -0.78
     anecdotal
    -0.77
     comparatively
    -0.76
     inclined
    -0.76
     avail
    -0.76
     defin
    -0.75
    POSITIVE LOGITS
     Skies
    1.31
     Wheels
    1.28
     Tracks
    1.24
     Walls
    1.23
     Mouth
    1.23
     Noise
    1.22
     Trees
    1.22
     Pieces
    1.21
     Madness
    1.21
     Life
    1.21
    Activations Density 0.219%

    No Known Activations