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. 1-RES-JB
    6. 1473
    Prev
    Next
    INDEX
    Explanations

    mentions of the word "triangle"

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "tri" and its variations, particularly related to triangles or triangular shapes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Canary
    -0.82
    manship
    -0.76
    icago
    -0.71
    ãģ®éŃĶ
    -0.69
    mson
    -0.64
    ãĤ¼ãĤ¦ãĤ¹
    -0.62
    INESS
    -0.61
     Kills
    -0.61
     Confederation
    -0.60
     [&
    -0.59
    POSITIVE LOGITS
    ple
    1.08
    plets
    0.99
    adic
    0.96
    umph
    0.94
    pling
    0.92
    pled
    0.91
    angle
    0.90
    urnal
    0.90
    tri
    0.88
    plet
    0.85
    Activations Density 0.004%

    No Known Activations