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. 1303
    Prev
    Next
    INDEX
    Explanations

    phrases related to political news and governmental actions, specifically referencing international relations and political tensions

    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
    +0.27
    1.0%
    1967
    +0.18
    0.7%
    50
    +0.13
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    16
    +0.27
    0.10
    1967
    +0.18
    0.08
    50
    +0.13
    0.07
    Negative Logits
    ypeł
    -0.65
     pensieri
    -0.60
     commenti
    -0.60
    @[+][
    -0.57
    idać
    -0.57
     messaggi
    -0.56
    ))^{
    -0.55
     gardien
    -0.54
     sentimenti
    -0.54
    drage
    -0.54
    POSITIVE LOGITS
     Perci
    0.74
     Áng
    0.73
     Apare
    0.71
     stiamo
    0.68
     Darío
    0.67
     Diez
    0.66
     Egli
    0.65
     Haci
    0.63
     vuol
    0.62
     pié
    0.59
    Activations Density 0.380%

    No Known Activations