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. 8556
    Prev
    Next
    INDEX
    Explanations

    Japanese names ending in "ichi"

    oai_token-act-pair · gpt-3.5-turbo

    repeated mentions of a specific name or entity, particularly focusing on the term "Ichi"

    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
     alleg
    -0.70
     campaigns
    -0.69
     pal
    -0.67
     Dove
    -0.66
     tears
    -0.64
     relative
    -0.63
     fore
    -0.63
     shadow
    -0.62
     hindsight
    -0.62
     yours
    -0.62
    POSITIVE LOGITS
    ichi
    4.52
    hei
    1.49
    itsu
    1.27
    ishi
    1.26
    ei
    1.25
    ich
    1.24
    ji
    1.23
    ataka
    1.22
    atsu
    1.22
    achi
    1.19
    Activations Density 0.009%

    No Known Activations