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. 787
    Prev
    Next
    INDEX
    Explanations

    words related to pools or pooling, including terms like "carpool" and "threadPool"

    oai_token-act-pair · gpt-3.5-turbo

    references to pools, specifically various types of pools and their contexts

    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
     dear
    -0.70
     Vis
    -0.68
     pronounce
    -0.65
     Flavoring
    -0.63
     leg
    -0.61
    ACTED
    -0.60
    ®
    -0.60
     Armenian
    -0.59
     indoctr
    -0.59
     Aure
    -0.59
    POSITIVE LOGITS
    pool
    4.86
    Pool
    2.33
     pools
    1.84
     Pool
    1.81
     pool
    1.79
     pooled
    1.08
    cloud
    0.99
    stream
    0.97
    storm
    0.95
    adder
    0.93
    Activations Density 0.011%

    No Known Activations