© 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. 15863
    Prev
    Next
    INDEX
    Explanations

    medical mechanisms and processes

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to processes or systems that operate to achieve specific outcomes

    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
    ear
    -0.69
    Beat
    -0.69
     Eps
    -0.68
    iyah
    -0.67
     bless
    -0.66
     Premiere
    -0.65
     Bold
    -0.64
    corn
    -0.64
    head
    -0.63
     Pride
    -0.62
    POSITIVE LOGITS
     mechanisms
    3.53
     mechanism
    2.81
     Mechan
    1.82
     levers
    1.64
     processes
    1.58
     pathways
    1.57
     circuitry
    1.40
     machinery
    1.35
     structures
    1.34
     procedures
    1.30
    Activations Density 0.030%

    No Known Activations