© 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 · Initial Gemma-2B-IT SAEs
    5. Gemma-2B-IT
    7. Residual Stream
    9. 12-RES-JB
    11. 13416
    Prev
    Next
    INDEX
    Explanations

    structural patterns related to code or programming, specifically focusing on syntax and expressions

    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
    +0.13
    0.4%
    1343
    +0.12
    0.3%
    126
    +0.07
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    691
    +0.13
    0.01
    1061
    +0.12
    0.02
    876
    +0.07
    -0.01
    Negative Logits
    Também
    -0.72
    Escuela
    -0.69
    Ultimo
    -0.68
    Primeiro
    -0.66
    Šaltiniai
    -0.66
    Agencia
    -0.65
    Boletín
    -0.63
    Marzo
    -0.63
    Fontes
    -0.61
    Enllaços
    -0.61
    POSITIVE LOGITS
     effe
    1.64
     increa
    1.51
     ?...
    1.50
     fuf
    1.49
     fto
    1.48
     purcha
    1.47
     unden
    1.45
     guarante
    1.43
     strick
    1.42
     squa
    1.42
    Activations Density 0.190%

    No Known Activations