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. 7416
    Prev
    Next
    INDEX
    Explanations

    phrases or expressions emphasizing a specific point or idea

    oai_token-act-pair · gpt-3.5-turbo

    references to considerations or contexts that influence actions or decisions

    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
    heses
    -0.80
    quer
    -0.74
    bugs
    -0.73
    late
    -0.73
    taboola
    -0.70
    ixels
    -0.70
    humans
    -0.69
    oons
    -0.69
    ori
    -0.67
    ombs
    -0.66
    POSITIVE LOGITS
     caveat
    1.12
     backdrop
    1.01
     disclaimer
    0.93
     newfound
    0.88
     knowledge
    0.88
     mindset
    0.84
     caveats
    0.83
     background
    0.79
     intention
    0.79
     pedigree
    0.78
    Activations Density 0.053%

    No Known Activations