© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Initial Gemma-2B-IT SAEs
    5. Gemma-2B-IT
    7. Residual Stream
    9. 12-RES-JB
    11. 12602
    Prev
    Next
    INDEX
    Explanations

    monetary amounts and numerical increases or decreases

    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
    453
    +0.09
    0.2%
    609
    +0.09
    0.2%
    694
    +0.08
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    609
    +0.09
    0.04
    453
    +0.09
    0.04
    1804
    +0.08
    0.03
    Negative Logits
     unspeak
    -1.44
     disagre
    -1.42
     reluct
    -1.42
     unve
    -1.39
     increa
    -1.39
     tolerably
    -1.37
     maneu
    -1.37
     encomp
    -1.36
     inev
    -1.34
     indestru
    -1.33
    POSITIVE LOGITS
    ↗
    0.78
    ↘
    0.58
    Geographie
    0.56
     maior
    0.52
    |=\
    0.52
    more
    0.50
    |+\
    0.50
     Процитовано
    0.48
    OleDb
    0.47
    maxcdn
    0.47
    Activations Density 0.144%

    No Known Activations