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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 5192
    Prev
    Next
    INDEX
    Explanations

    phrases indicating a need to improve or fix something in the context of sports

    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
    1928
    +0.07
    0.2%
    674
    +0.07
    0.2%
    1253
    +0.07
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    2045
    +0.07
    0.04
    162
    +0.07
    0.02
    1269
    +0.07
    0.03
    Negative Logits
     effe
    -1.38
     ftu
    -1.33
     aen
    -1.31
     „,
    -1.27
     Juf
    -1.26
     fta
    -1.26
     NOO
    -1.26
     overla
    -1.25
     inev
    -1.23
     increa
    -1.22
    POSITIVE LOGITS
    ContentAsync
    0.63
    álbum
    0.61
    脚注の使い方
    0.61
    MemoryWarning
    0.61
     miracle
    0.60
     luck
    0.60
     prove
    0.58
    🤞
    0.58
    IsContent
    0.57
     win
    0.56
    Activations Density 0.259%

    No Known Activations