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. 1524
    Prev
    Next
    INDEX
    Explanations

    timestamps in a specific format

    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
    1343
    +0.21
    0.6%
    674
    +0.12
    0.4%
    453
    +0.08
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1343
    +0.21
    0.04
    147
    +0.12
    0.02
    247
    +0.08
    0.03
    Negative Logits
     unspeak
    -0.85
     vainly
    -0.69
     shenan
    -0.67
     indescri
    -0.67
     apprehen
    -0.66
     gaily
    -0.66
     horrend
    -0.64
     rascal
    -0.64
     nobly
    -0.63
     luxuriant
    -0.63
    POSITIVE LOGITS
     utop
    1.03
     alkoh
    1.03
     solidar
    1.02
     reger
    1.01
     geolog
    0.98
     asfal
    0.96
     uhr
    0.95
     vitale
    0.94
     lapto
    0.94
     spion
    0.94
    Activations Density 0.065%

    No Known Activations