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. 23998
    Prev
    Next
    INDEX
    Explanations

    terms related to the word "mantle"

    oai_token-act-pair · gpt-3.5-turbo

    mentions of specific names or terminology related to technology and political correctness

    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
     elig
    -0.76
     Pon
    -0.72
     registered
    -0.72
     Aman
    -0.69
     Alto
    -0.68
     AUTH
    -0.66
     Hawaiian
    -0.65
     altern
    -0.65
     ELECT
    -0.65
     Yose
    -0.64
    POSITIVE LOGITS
    antle
    2.93
    ifle
    1.38
    ordon
    1.31
    pox
    1.29
     cull
    1.24
     correctness
    1.16
    ifling
    1.15
    opard
    1.12
     Dylan
    1.11
    ifles
    1.07
    Activations Density 0.047%

    No Known Activations