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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    2. Joseph Bloom · Initial Gemma-2B-IT SAEs
    3. Gemma-2B-IT
    4. Residual Stream
    5. 12-RES-JB
    6. 7767
    Prev
    Next
    INDEX
    Explanations

    phrases referring to reasons or justifications

    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
    468
    +0.09
    0.2%
    1499
    +0.08
    0.2%
    1438
    +0.07
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1908
    +0.09
    0.04
    1056
    +0.08
    0.05
    1801
    +0.07
    0.03
    Negative Logits
     :)))
    -0.62
     pymysql
    -0.61
     nutella
    -0.59
     tiffany
    -0.59
     macrop
    -0.58
     michelin
    -0.58
     cryst
    -0.57
     bonbons
    -0.57
     ecru
    -0.56
     tricot
    -0.56
    POSITIVE LOGITS
     Fö
    0.57
    Ár
    0.51
     Palembang
    0.51
     induk
    0.51
     Jä
    0.50
     roh
    0.50
     considération
    0.50
    脚注の使い方
    0.50
     Ukraina
    0.49
     sarili
    0.49
    Activations Density 0.455%

    No Known Activations