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. 10101
    Prev
    Next
    INDEX
    Explanations

    Phrases related to programming practices and software development techniques

    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
    876
    +0.27
    1.1%
    1577
    +0.21
    0.8%
    50
    +0.17
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    876
    +0.27
    -0.03
    678
    +0.21
    0.17
    2044
    +0.17
    0.19
    Negative Logits
     catedral
    -0.87
     Demokrat
    -0.86
     Jerusalén
    -0.81
    Muerte
    -0.79
    Él
    -0.79
    الدراسه
    -0.77
    jupiter
    -0.77
    Perú
    -0.77
    Educación
    -0.76
    TagMode
    -0.76
    POSITIVE LOGITS
     impra
    2.47
     encomp
    2.42
     resear
    2.39
     maneu
    2.38
     increa
    2.35
     affor
    2.31
     inev
    2.28
     Intere
    2.24
     practition
    2.23
     uninten
    2.22
    Activations Density 4.764%

    No Known Activations