© 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. 4089
    Prev
    Next
    INDEX
    Explanations

    instances of the word "processed."

    oai_token-act-pair · gpt-3.5-turbo

    mentions of processed foods and their health implications

    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
     Powers
    -0.76
    ettings
    -0.73
    lihood
    -0.72
     Pact
    -0.70
    ods
    -0.70
     Talent
    -0.70
    ilities
    -0.69
     Writers
    -0.69
     Dev
    -0.68
    pring
    -0.67
    POSITIVE LOGITS
     processed
    3.57
     packaged
    1.36
     refined
    1.34
     scanned
    1.33
     filtered
    1.29
     parsed
    1.27
     retrieved
    1.23
     harvested
    1.23
     saturated
    1.20
     analyzed
    1.18
    Activations Density 0.016%

    No Known Activations