© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 3-RES-JB
    11. 20647
    Prev
    Next
    INDEX
    Explanations

    words related to official terms and designations, often associated with bureaucratic or formal language

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "ations" in various contexts

    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
    Effective
    -0.63
    awar
    -0.63
    CHAR
    -0.58
     forestry
    -0.57
    cod
    -0.56
    asus
    -0.54
    erton
    -0.52
    URRENT
    -0.52
    thia
    -0.52
     motion
    -0.52
    POSITIVE LOGITS
    ettings
    1.26
    hip
    1.25
    uggest
    1.17
    hower
    1.12
    hips
    1.10
    poons
    1.09
    leeve
    1.07
    hooting
    1.03
    uits
    1.03
    ystem
    1.00
    Activations Density 0.077%

    No Known Activations