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. 3540
    Prev
    Next
    INDEX
    Explanations

    keywords related to data processing instructions and input information

    oai_token-act-pair · gpt-3.5-turbo

    instances of the term "input" in various contexts

    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
     Seraph
    -0.72
    astern
    -0.65
    neys
    -0.65
    itect
    -0.62
    adder
    -0.61
    llan
    -0.61
    esa
    -0.60
    ETA
    -0.60
    ISS
    -0.59
     Apart
    -0.58
    POSITIVE LOGITS
    ted
    1.06
    enza
    0.93
     inputs
    0.92
     input
    0.89
    ting
    0.87
    ters
    0.85
     "$:/
    0.76
    ierrez
    0.76
    input
    0.75
     guiName
    0.74
    Activations Density 0.010%

    No Known Activations