© 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. 19752
    Prev
    Next
    INDEX
    Explanations

    proper names or entities

    oai_token-act-pair · gpt-3.5-turbo

    prominent nouns and proper nouns related to significant topics or figures

    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
     Gil
    -0.67
    )--
    -0.61
    udo
    -0.60
     LIFE
    -0.60
    senal
    -0.58
    Miami
    -0.58
    utterstock
    -0.58
     adolesc
    -0.58
    Ĥ¬
    -0.58
    ength
    -0.56
    POSITIVE LOGITS
     hadn
    0.72
     shouldn
    0.64
     deserved
    0.64
    mite
    0.64
    bably
    0.63
     should
    0.62
     couldn
    0.62
    seless
    0.62
    mon
    0.61
    hematically
    0.61
    Activations Density 0.898%

    No Known Activations