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. 11-RES-JB
    6. 3047
    Prev
    Next
    INDEX
    Explanations

    phrases related to contrasting viewpoints and discussions

    oai_token-act-pair · gpt-3.5-turbo

    instances of punctuation, particularly commas and periods, indicating a focus on sentence structure

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ãĤ¨ãĥ«
    -0.80
    èĢħ
    -0.76
    ļéĨĴ
    -0.73
    ãĤ¼ãĤ¦ãĤ¹
    -0.72
    erenn
    -0.70
    ROR
    -0.68
    iku
    -0.68
    abama
    -0.68
    wash
    -0.67
    lass
    -0.64
    POSITIVE LOGITS
     nevertheless
    1.14
     there
    1.00
     it
    0.94
     nonetheless
    0.93
     namely
    0.90
    there
    0.85
     albeit
    0.82
     neither
    0.80
     they
    0.77
     suffice
    0.76
    Activations Density 0.184%

    No Known Activations