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. 4-RES-JB
    6. 3790
    Prev
    Next
    INDEX
    Explanations

    phrases that mention or imply a specific purpose

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "purpose."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 4-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    4
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Wolves
    -0.67
     Roses
    -0.65
     tears
    -0.63
     Whitman
    -0.62
    omez
    -0.62
     nerves
    -0.61
     Temper
    -0.61
     Thumbnails
    -0.61
    etts
    -0.61
    thumbnails
    -0.60
    POSITIVE LOGITS
    ful
    1.17
    lessly
    1.09
    lessness
    0.96
    fully
    0.94
    fulness
    0.94
    ality
    0.89
    tenance
    0.80
    folios
    0.80
    eous
    0.80
    fare
    0.79
    Activations Density 0.012%

    No Known Activations