© 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
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 0-RES-JB
    6. 12055
    Prev
    Next
    INDEX
    Explanations

    terms related to loops and local communities

    oai_token-act-pair · gpt-3.5-turbo

    references to loops and local contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     bow
    -0.76
     stomp
    -0.76
     tee
    -0.72
     inches
    -0.70
     Targ
    -0.70
     sermon
    -0.69
     toe
    -0.68
     pict
    -0.68
     Din
    -0.67
     wre
    -0.66
    POSITIVE LOGITS
    loop
    3.28
    orbit
    1.87
    local
    1.72
    liquid
    1.46
    cigarettes
    1.34
    cigarette
    1.23
    rc
    1.21
    lab
    1.19
    bucks
    1.19
    communication
    1.19
    Activations Density 0.054%

    No Known Activations