© 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. 3-RES-JB
    6. 9676
    Prev
    Next
    INDEX
    Explanations

    words related to medical conditions such as "euphoria" and "glutamate."

    oai_token-act-pair · gpt-3.5-turbo

    words related to observations, details, and classifications

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     redistributed
    -0.52
    milo
    -0.47
     enriched
    -0.45
     pestic
    -0.45
     indexed
    -0.45
     cradle
    -0.45
     triv
    -0.44
     predec
    -0.43
     unrestricted
    -0.43
    ciating
    -0.43
    POSITIVE LOGITS
    lihood
    0.61
    ï¸ı
    0.55
    BACK
    0.44
    puff
    0.44
    ble
    0.43
     Pwr
    0.43
    ®
    0.42
    back
    0.42
    hole
    0.41
    lde
    0.41
    Activations Density 0.712%

    No Known Activations