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. 3-RES-JB
    6. 19779
    Prev
    Next
    INDEX
    Explanations

    verbs that indicate action or intention

    oai_token-act-pair · gpt-3.5-turbo

    action verbs indicating realization or achievement

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ItemTracker
    -0.90
    pmwiki
    -0.81
    lihood
    -0.63
    pie
    -0.63
     pie
    -0.60
     gonna
    -0.59
     To
    -0.59
    ammy
    -0.58
    imaru
    -0.58
     knack
    -0.57
    POSITIVE LOGITS
    ISE
    0.73
    inton
    0.67
    WT
    0.67
    ilee
    0.66
    uke
    0.61
    Cre
    0.59
    VIDIA
    0.58
     Ratio
    0.57
    Ü
    0.57
    cliffe
    0.57
    Activations Density 0.589%

    No Known Activations