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. 16047
    Prev
    Next
    INDEX
    Explanations

    technical and engineering terms related to structures and systems

    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
    1870
    +0.16
    0.5%
    604
    +0.10
    0.3%
    939
    +0.10
    0.3%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1870
    +0.16
    0.04
    1672
    +0.10
    0.04
    1531
    +0.10
    0.04
    Negative Logits
     vieilles
    -0.82
     célé
    -0.81
     stefan
    -0.81
     clôture
    -0.79
     superbes
    -0.71
     dépens
    -0.70
     véritables
    -0.69
     vrais
    -0.69
     ensu
    -0.68
     karl
    -0.68
    POSITIVE LOGITS
     problems
    0.69
     issues
    0.68
     pymysql
    0.67
     nepiecieš
    0.60
     smtplib
    0.57
     challenges
    0.57
     gardent
    0.57
     complications
    0.56
     loss
    0.56
     violations
    0.54
    Activations Density 0.401%

    No Known Activations