© 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. 0-RES-JB
    11. 11393
    Prev
    Next
    INDEX
    Explanations

    formal language negations

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "didn't" or its variations in different contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     attain
    -0.62
    alg
    -0.62
    ipes
    -0.60
    cot
    -0.53
     Article
    -0.52
     approximately
    -0.52
     PUBLIC
    -0.52
     reciprocal
    -0.51
     ens
    -0.51
     instances
    -0.50
    POSITIVE LOGITS
     didn
    2.94
     hadn
    2.53
    didn
    2.51
     wasn
    2.38
     weren
    2.32
     couldn
    2.29
     didnt
    2.29
     Didn
    2.21
     wouldn
    2.10
     hasn
    1.98
    Activations Density 0.033%

    No Known Activations