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. 7248
    Prev
    Next
    INDEX
    Explanations

    proper nouns or specific labels

    oai_token-act-pair · gpt-3.5-turbo

    sentences that begin with "The" or contain notable data or statistics highlighted by "The"

    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
    udic
    -0.70
    rib
    -0.69
    sonian
    -0.67
    morrow
    -0.65
    ofi
    -0.65
    iev
    -0.65
     someday
    -0.64
    ienne
    -0.63
    fab
    -0.63
    annels
    -0.63
    POSITIVE LOGITS
     median
    1.42
     percentage
    1.40
     average
    1.31
     ratio
    1.27
     percentages
    1.24
     proportion
    1.20
     disparity
    1.17
     number
    1.15
     lowest
    1.15
     discrepancy
    1.14
    Activations Density 0.233%

    No Known Activations