© 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. 24460
    Prev
    Next
    INDEX
    Explanations

    names of locations or people with the suffix "und" or similar sounds

    oai_token-act-pair · gpt-3.5-turbo

    suffixes and letter combinations that indicate plural or participial forms in words

    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
     constitu
    -0.85
    natureconservancy
    -0.85
    é¾
    -0.80
    Preview
    -0.78
    SpaceEngineers
    -0.78
     Archdemon
    -0.77
    £ı
    -0.76
     Cosponsors
    -0.75
    TAG
    -0.75
    Ń·
    -0.74
    POSITIVE LOGITS
    shire
    0.79
    estone
    0.69
    raft
    0.68
    ja
    0.68
     Jenner
    0.68
    abase
    0.67
    ishes
    0.67
     Elise
    0.66
    loe
    0.66
    ijk
    0.64
    Activations Density 0.182%

    No Known Activations