Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 24274
    Prev
    Next
    INDEX
    Explanations

    highway names with corresponding numbers

    oai_token-act-pair · gpt-3.5-turbo

    negative prefixes or descriptors associated with roads and highways

    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
     scalp
    -0.58
     free
    -0.58
     reign
    -0.58
     bailed
    -0.58
     Columbia
    -0.58
     reload
    -0.55
     nights
    -0.54
     correction
    -0.53
     bur
    -0.53
     whichever
    -0.52
    POSITIVE LOGITS
    acan
    1.04
    annis
    0.94
    aca
    0.90
    é¾įå
    0.82
    olit
    0.80
    uct
    0.80
    heart
    0.80
    umerable
    0.79
    ACA
    0.78
    Glass
    0.75
    Activations Density 0.091%

    No Known Activations