© 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
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 0-RES-JB
    6. 9480
    Prev
    Next
    INDEX
    Explanations

    phrases related to remixing or editing content

    oai_token-act-pair · gpt-3.5-turbo

    references to musical remixes and cultural appropriation

    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
     Radar
    -0.73
    iveness
    -0.68
     Hau
    -0.68
     Dire
    -0.68
     Tide
    -0.68
    BILITY
    -0.67
     Fein
    -0.67
    ieving
    -0.66
     Coh
    -0.66
    naires
    -0.66
    POSITIVE LOGITS
     remix
    2.50
     Remix
    2.43
     mash
    2.10
     appropriation
    1.59
     takedown
    1.37
     Mash
    1.32
    rito
    1.31
     parody
    1.24
     appropriated
    1.17
     mont
    1.15
    Activations Density 0.034%

    No Known Activations