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. 5728
    Prev
    Next
    INDEX
    Explanations

    words related to importing

    oai_token-act-pair · gpt-3.5-turbo

    keywords related to importing data or resources

    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
    Winged
    -0.74
    hered
    -0.74
     Surv
    -0.71
     Chel
    -0.65
     Hastings
    -0.65
     Heroic
    -0.65
     Rai
    -0.64
    Ward
    -0.64
     Kara
    -0.64
    bed
    -0.64
    POSITIVE LOGITS
     import
    3.84
    import
    2.86
     importing
    2.64
     imports
    2.57
     Import
    2.31
     export
    2.28
     imported
    2.19
    Import
    2.14
     exporting
    1.82
    export
    1.76
    Activations Density 0.027%

    No Known Activations