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. 10006
    Prev
    Next
    INDEX
    Explanations

    steps or instructions in a recipe

    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
    2034
    +0.16
    0.5%
    2019
    +0.15
    0.5%
    752
    +0.11
    0.3%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    946
    +0.16
    0.03
    2019
    +0.15
    0.03
    463
    +0.11
    0.02
    Negative Logits
     pamph
    -1.33
     Czechos
    -1.26
     Abbé
    -1.15
     Souha
    -1.15
     viciss
    -1.15
     unwarran
    -1.14
     accla
    -1.13
     monaster
    -1.13
     disagre
    -1.12
     År
    -1.11
    POSITIVE LOGITS
     Then
    0.82
    Then
    0.76
     rechange
    0.69
    HttpDelete
    0.66
    then
    0.66
    Optional
    0.65
     proceed
    0.64
     then
    0.64
     transfert
    0.63
    المصادر
    0.62
    Activations Density 0.148%

    No Known Activations