Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 11-RES-JB
    6. 15096
    Prev
    Next
    INDEX
    Explanations

    French words and phrases

    oai_token-act-pair · gpt-3.5-turbo

    instances of the letter 'é' in various contexts, particularly in relation to the French language

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Pledge
    -0.65
     bulls
    -0.65
     Shining
    -0.64
     Parenthood
    -0.64
     hover
    -0.64
     sorely
    -0.63
     Curse
    -0.60
     assembly
    -0.60
     Abu
    -0.59
     Cobra
    -0.59
    POSITIVE LOGITS
    lé
    1.30
    ré
    1.21
    mie
    1.20
    lite
    1.10
    rie
    1.08
    rique
    1.06
    cé
    1.05
    conom
    0.97
    vez
    0.97
    dit
    0.96
    Activations Density 0.035%

    No Known Activations