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. 3227
    Prev
    Next
    INDEX
    Explanations

    references to a specific place or location containing the string "ovo"

    oai_token-act-pair · gpt-3.5-turbo

    references to a specific location or region, particularly "Bregovo."

    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
     Dee
    -0.72
     bats
    -0.71
     Babe
    -0.70
     Charlotte
    -0.68
     pitchers
    -0.68
     angels
    -0.64
     batters
    -0.63
    casts
    -0.63
     Katrina
    -0.63
     Stard
    -0.62
    POSITIVE LOGITS
    ovo
    5.18
    ovi
    1.14
    ova
    1.13
    ventus
    1.08
    ovic
    1.04
     Ukrain
    1.04
    ansk
    1.04
    orsi
    0.99
    achev
    0.98
    ove
    0.97
    Activations Density 0.024%

    No Known Activations