© 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. 22382
    Prev
    Next
    INDEX
    Explanations

    phrases related to the action of swallowing

    oai_token-act-pair · gpt-3.5-turbo

    references to acts of swallowing or related digestive actions

    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
     Omar
    -0.80
     profiling
    -0.76
    modules
    -0.74
     Mod
    -0.73
    Mas
    -0.67
     SOC
    -0.65
     Morales
    -0.65
    Trend
    -0.65
    rity
    -0.63
     Mosque
    -0.62
    POSITIVE LOGITS
     swallow
    3.66
     swallowed
    3.16
     swallowing
    3.07
     swall
    2.87
     chew
    1.45
     chewing
    1.34
     ingested
    1.34
     gob
    1.29
     sucking
    1.20
     choking
    1.19
    Activations Density 0.023%

    No Known Activations