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. 15665
    Prev
    Next
    INDEX
    Explanations

    technical terms and descriptions related to hardware, technology, and mechanical processes

    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
    +0.22
    0.7%
    1967
    +0.20
    0.6%
    1385
    +0.15
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1967
    +0.22
    0.05
    562
    +0.20
    0.04
    1635
    +0.15
    0.03
    Negative Logits
     tremb
    -0.80
     chande
    -0.75
     bourgeo
    -0.75
    sieur
    -0.74
     ladri
    -0.74
     Messieurs
    -0.72
     peppa
    -0.71
     germain
    -0.70
     fuo
    -0.70
     milf
    -0.70
    POSITIVE LOGITS
    Índice
    0.56
    Descubre
    0.51
    Linki
    0.50
    Már
    0.49
    Frankreich
    0.48
     autorytatywna
    0.47
    Kör
    0.47
    Morfologia
    0.47
    Encontra
    0.46
    Noter
    0.46
    Activations Density 0.222%

    No Known Activations