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. 0-RES-JB
    6. 14468
    Prev
    Next
    INDEX
    Explanations

    symbols representing structural elements, such as slashes or vertical bars

    oai_token-act-pair · gpt-3.5-turbo

    symbols or characters that could represent formatting or structural elements in text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Regulatory
    -0.72
     Revised
    -0.65
     Related
    -0.64
     CrossRef
    -0.63
    ffen
    -0.61
     Problems
    -0.59
     Riders
    -0.59
    chen
    -0.59
     Cavern
    -0.59
    isite
    -0.57
    POSITIVE LOGITS
    \/
    1.91
     much
    0.88
    inventoryQuantity
    0.81
    retty
    0.80
    much
    0.79
    oji
    0.74
     MUCH
    0.73
     Flask
    0.72
    å§«
    0.70
     sidx
    0.68
    Activations Density 0.000%

    No Known Activations