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. 2985
    Prev
    Next
    INDEX
    Explanations

    mentions of specific quantities or statistics

    oai_token-act-pair · gpt-3.5-turbo

    references to specific periods or instances in economic or fiscal reporting

    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
    igans
    -0.81
    ular
    -0.60
    uthor
    -0.56
    ÃŃ
    -0.56
    âĶģ
    -0.56
    Vo
    -0.56
     someday
    -0.54
    ican
    -0.53
    Vector
    -0.53
     appe
    -0.53
    POSITIVE LOGITS
     same
    1.17
     past
    1.11
     preceding
    1.08
     fourth
    1.04
     first
    1.03
     period
    1.03
     year
    1.02
     month
    1.02
     second
    0.99
     latest
    0.97
    Activations Density 0.127%

    No Known Activations