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. 3965
    Prev
    Next
    INDEX
    Explanations

    instances where the term "Seven" is mentioned

    oai_token-act-pair · gpt-3.5-turbo

    references to the number "seven" or phrases that incorporate the concept of "seven."

    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
    yg
    -0.72
     ali
    -0.64
    utm
    -0.63
     propag
    -0.63
    amar
    -0.63
    dyl
    -0.62
    OG
    -0.61
    ags
    -0.60
     layer
    -0.60
    call
    -0.59
    POSITIVE LOGITS
     Seven
    3.56
    Seven
    2.31
     Nine
    1.88
     Six
    1.84
     Eleven
    1.82
     Eight
    1.78
     Five
    1.74
     Ten
    1.60
     Thirty
    1.57
     Twelve
    1.54
    Activations Density 0.010%

    No Known Activations