© 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. 4-RES-JB
    11. 20124
    Prev
    Next
    INDEX
    Explanations

    features or modes that allow users to perform specific actions within different software applications

    oai_token-act-pair · gpt-3.5-turbo

    actions or functionalities provided by applications or software

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

    No Comments

    Negative Logits
    querque
    -0.74
    Ô
    -0.67
    bage
    -0.66
     Beasts
    -0.65
    natureconservancy
    -0.65
    venant
    -0.65
    xual
    -0.65
     millenn
    -0.64
    PLIED
    -0.64
    lihood
    -0.63
    POSITIVE LOGITS
    tered
    1.01
    tering
    0.91
     us
    0.82
    icia
    0.81
    hey
    0.79
    heet
    0.78
    iton
    0.75
    isters
    0.75
    selves
    0.72
    rack
    0.70
    Activations Density 0.019%

    No Known Activations