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. 3-RES-JB
    6. 2524
    Prev
    Next
    INDEX
    Explanations

    political and technical terms

    oai_token-act-pair · gpt-3.5-turbo

    content related to analysis and critique of media and societal issues

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    rower
    -0.65
    etsk
    -0.64
    agonist
    -0.60
    Downloadha
    -0.59
    aird
    -0.59
    xon
    -0.58
    viks
    -0.56
     Seah
    -0.55
    orsche
    -0.54
    uncle
    -0.54
    POSITIVE LOGITS
     donors
    0.58
    bugs
    0.54
     purposes
    0.53
     levels
    0.52
     footing
    0.52
     sites
    0.51
     networking
    0.51
     materials
    0.51
     donations
    0.50
     emergencies
    0.49
    Activations Density 2.452%

    No Known Activations