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. 2889
    Prev
    Next
    INDEX
    Explanations

    references to historical events and legal terms

    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
    1741
    +0.15
    0.5%
    1177
    +0.14
    0.4%
    1535
    +0.11
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    678
    +0.15
    0.06
    1671
    +0.14
    0.06
    1430
    +0.11
    0.05
    Negative Logits
    <bos>
    -0.82
     이미
    -0.72
     виправи
    -0.62
     потім
    -0.61
     
    -0.61
     العزيز
    -0.60
    addMouseListener
    -0.60
    以及
    -0.60
    使得
    -0.59
      
    -0.59
    POSITIVE LOGITS
     igno
    2.02
     wien
    1.92
     pessi
    1.92
     milano
    1.88
     accla
    1.87
     mef
    1.87
     hcm
    1.86
     embra
    1.85
     effe
    1.84
     chande
    1.83
    Activations Density 0.574%

    No Known Activations