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. 1-RES-JB
    6. 1328
    Prev
    Next
    INDEX
    Explanations

    phrases related to specific popular culture references, such as "Half-Life"

    oai_token-act-pair · gpt-3.5-turbo

    references to the phrase "Half" along with related terms

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    HCR
    -0.80
    Trend
    -0.73
    à¼
    -0.68
    ãģķ
    -0.64
     Nanto
    -0.64
    è¦ļéĨĴ
    -0.63
    orship
    -0.63
     electronically
    -0.62
    ISE
    -0.62
    essional
    -0.62
    POSITIVE LOGITS
     Half
    0.92
    eteen
    0.89
    pipe
    0.84
    Half
    0.84
    terness
    0.83
    heartedly
    0.82
    quarters
    0.76
    eworks
    0.74
    regate
    0.72
    bye
    0.72
    Activations Density 0.006%

    No Known Activations