© 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. 8074
    Prev
    Next
    INDEX
    Explanations

    mentions of activities or objects associated with meal preparation or cooking

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "meal" and its variations

    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
     Prof
    -0.65
     psychiatrist
    -0.65
     Idlib
    -0.65
     Bosnia
    -0.65
     Cant
    -0.65
     DK
    -0.65
     no
    -0.64
     notebooks
    -0.63
     Sass
    -0.62
     MX
    -0.61
    POSITIVE LOGITS
    eal
    4.85
    eals
    2.21
    ealing
    1.93
    ealous
    1.28
    egu
    1.20
    ream
    1.16
    ell
    1.08
    eat
    1.08
    ey
    1.07
    eed
    1.05
    Activations Density 0.009%

    No Known Activations