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. 11-RES-JB
    6. 4421
    Prev
    Next
    INDEX
    Explanations

    numeric comparisons and quantities

    oai_token-act-pair · gpt-3.5-turbo

    comparative statistics and numerical values related to demographics or health

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    RAFT
    -0.61
     Soon
    -0.61
    ãĥŁ
    -0.58
    hyde
    -0.58
    acity
    -0.58
    Soon
    -0.55
    ota
    -0.55
    birds
    -0.54
    Contribut
    -0.54
     Contributions
    -0.54
    POSITIVE LOGITS
     double
    1.27
     triple
    1.19
     twice
    1.16
     quadru
    1.06
     tripled
    1.06
     doubled
    1.05
     doubling
    1.00
     doubles
    0.98
     half
    0.97
     hal
    0.89
    Activations Density 0.114%

    No Known Activations