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. 22144
    Prev
    Next
    INDEX
    Explanations

    words related to chemical compounds

    oai_token-act-pair · gpt-3.5-turbo

    adjectives that describe qualities or experiences

    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
     Apprentice
    -0.66
    士
    -0.65
     Accountability
    -0.63
     IPA
    -0.63
     IPM
    -0.62
     chef
    -0.61
     budget
    -0.61
     volunteered
    -0.60
     ADA
    -0.60
     Keller
    -0.59
    POSITIVE LOGITS
    rous
    4.77
    rious
    1.31
    rouse
    1.27
    rus
    1.19
    ries
    1.16
    rogen
    1.09
    rou
    1.06
    rium
    1.05
    ric
    1.01
    ry
    0.99
    Activations Density 0.017%

    No Known Activations