© 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
    2. Joseph Bloom · Initial Gemma-2B-IT SAEs
    3. Gemma-2B-IT
    4. Residual Stream
    5. 12-RES-JB
    6. 11834
    Prev
    Next
    INDEX
    Explanations

    statements starting with "I think" or similar phrases indicating personal opinion or analysis

    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
    184
    +0.20
    0.7%
    468
    +0.17
    0.5%
    604
    +0.11
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    599
    +0.20
    0.05
    468
    +0.17
    0.04
    184
    +0.11
    0.02
    Negative Logits
     impra
    -1.83
     indestru
    -1.78
     disagre
    -1.71
     suscep
    -1.70
     reluct
    -1.66
     inappro
    -1.65
     disreg
    -1.60
     uninten
    -1.59
     increa
    -1.58
     unspeak
    -1.58
    POSITIVE LOGITS
     [
    0.70
    ,"
    0.69
    ,”
    0.68
    XMLSchema
    0.66
     يتيمه
    0.65
     ideolog
    0.64
    Tē
    0.63
    MessageOf
    0.62
     Walkover
    0.62
     typelib
    0.60
    Activations Density 0.219%

    No Known Activations