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 · Initial Gemma-2B-IT SAEs
    3. Gemma-2B-IT
    4. Residual Stream
    5. 12-RES-JB
    6. 403
    Prev
    Next
    INDEX
    Explanations

    phrases related to introspection and self-reflection

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    jbloom/Gemma-2b-IT-Residual-Stream-SAEs/gemma_2b_it_blocks.12.hook_resid_post_16384
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    HuggingFaceFW/fineweb
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.12.hook_resid_post
    Hook Layer
    12
    Architecture
    standard
    Context Size
    1,024
    Dataset
    Skylion007/openwebtext
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    478
    +0.19
    0.8%
    1806
    +0.17
    0.7%
    812
    +0.15
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    920
    +0.19
    0.07
    658
    +0.17
    0.08
    1806
    +0.15
    0.07
    Negative Logits
     isoli
    -0.94
     alpes
    -0.91
     akut
    -0.91
     kade
    -0.88
     monaster
    -0.88
     gmbh
    -0.85
     karton
    -0.85
     moza
    -0.85
     baum
    -0.85
     antik
    -0.85
    POSITIVE LOGITS
     gonna
    0.82
    re
    0.65
    muito
    0.65
    Estou
    0.64
    Horário
    0.64
     aren
    0.62
    外部链接
    0.60
    Gonna
    0.60
     كومونز
    0.60
     fiquei
    0.59
    Activations Density 0.172%

    No Known Activations