Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 14076
    Prev
    Next
    INDEX
    Explanations

    references to famous artists, particularly Henri Matisse

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "isse" in various forms

    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
     modem
    -0.67
     Bills
    -0.63
     amps
    -0.62
     recorder
    -0.62
     podcast
    -0.61
     conferences
    -0.59
    boat
    -0.59
     appearing
    -0.58
     recording
    -0.58
     conference
    -0.58
    POSITIVE LOGITS
    isse
    5.16
    iss
    1.61
    issance
    1.29
    izen
    1.11
    issa
    1.11
    itte
    0.98
    isen
    0.97
    isan
    0.96
    isson
    0.94
    ï
    0.93
    Activations Density 0.016%

    No Known Activations