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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 22209
    Prev
    Next
    INDEX
    Explanations

    mentions of the word “youngest”

    oai_token-act-pair · gpt-3.5-turbo

    references to age, specifically the youngest individuals in various 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
    fe
    -0.67
    works
    -0.63
     similar
    -0.63
    embed
    -0.62
     redirect
    -0.60
     neutral
    -0.60
    ject
    -0.57
     User
    -0.56
     guarantees
    -0.56
    fab
    -0.55
    POSITIVE LOGITS
     youngest
    3.75
     eldest
    2.77
     oldest
    2.40
     younger
    1.81
    young
    1.57
     tallest
    1.57
     newest
    1.54
     Younger
    1.54
     daughters
    1.50
     smallest
    1.49
    Activations Density 0.007%

    No Known Activations