© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 0-RES-JB
    11. 1635
    Prev
    Next
    INDEX
    Explanations

    references to locations or entities named "Italia"

    oai_token-act-pair · gpt-3.5-turbo

    references to "Italia" or related terms indicating Italy

    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
     push
    -0.72
     fart
    -0.69
     QB
    -0.67
     WIN
    -0.66
     thinker
    -0.65
     multiplier
    -0.64
     TB
    -0.62
     Tick
    -0.62
     MB
    -0.61
     quarterback
    -0.61
    POSITIVE LOGITS
    alia
    5.06
    alian
    1.99
    ali
    1.45
    ales
    1.28
    alis
    1.27
    elia
    1.25
    alias
    1.23
    uria
    1.18
    aly
    1.16
    alist
    1.16
    Activations Density 0.019%

    No Known Activations