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. 10722
    Prev
    Next
    INDEX
    Explanations

    terms related to suites or rooms in a hotel or office setting

    oai_token-act-pair · gpt-3.5-turbo

    references to specific software or suites

    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
    ],[
    -0.70
    IGN
    -0.68
    acements
    -0.68
    mate
    -0.64
    itory
    -0.64
     correction
    -0.63
    Correction
    -0.63
     Computing
    -0.62
    ãĥĦ
    -0.62
    ifix
    -0.61
    POSITIVE LOGITS
    renches
    0.95
    odore
    0.82
    achelor
    0.77
    20439
    0.73
    andum
    0.73
    Downloadha
    0.70
    rawdownloadcloneembedreportprint
    0.70
     logs
    0.70
    MSN
    0.69
     scraps
    0.68
    Activations Density 0.000%

    No Known Activations