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. 6729
    Prev
    Next
    INDEX
    Explanations

    references to unnamed or unidentified entities

    oai_token-act-pair · gpt-3.5-turbo

    references to unidentified or unnamed individuals and sources

    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
    achine
    -0.78
    akings
    -0.77
    lov
    -0.77
    Interstitial
    -0.75
    cdn
    -0.72
    otto
    -0.71
    annis
    -0.68
    ournals
    -0.67
    =-=-=-=-=-=-=-=-
    -0.66
    arium
    -0.65
    POSITIVE LOGITS
     unnamed
    1.01
     narrator
    0.76
     sourcing
    0.74
    REDACTED
    0.72
     anonymous
    0.71
     bystand
    0.70
    aneous
    0.69
     unidentified
    0.69
     sources
    0.68
     persons
    0.67
    Activations Density 0.011%

    No Known Activations