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. 10473
    Prev
    Next
    INDEX
    Explanations

    phrases including the unicode character sequence "âĢĵ"

    oai_token-act-pair · gpt-3.5-turbo

    the recurring character or symbol in the document

    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
    ysis
    -0.71
    ãĥ¼ãĥĨãĤ£
    -0.69
     Blu
    -0.68
    zzy
    -0.67
     redes
    -0.66
    uers
    -0.66
    hess
    -0.65
     parted
    -0.65
    iple
    -0.65
    oller
    -0.65
    POSITIVE LOGITS
    cases
    0.93
    ––
    0.89
    style
    0.88
    specific
    0.88
    micro
    0.88
    advertising
    0.85
    mediated
    0.83
    double
    0.82
    coll
    0.79
    rooms
    0.78
    Activations Density 0.029%

    No Known Activations