© 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. 4421
    Prev
    Next
    INDEX
    Explanations

    adjectives related to truth and understanding

    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
    872
    +0.13
    0.4%
    1253
    +0.08
    0.2%
    89
    +0.07
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    872
    +0.13
    0.07
    972
    +0.08
    0.03
    627
    +0.07
    0.03
    Negative Logits
     tew
    -0.90
     aen
    -0.90
     opport
    -0.90
     reluct
    -0.89
     depic
    -0.89
     aton
    -0.87
     encomp
    -0.85
     wien
    -0.85
     fath
    -0.85
     „,
    -0.84
    POSITIVE LOGITS
     fact
    0.66
     facts
    0.65
    calciatore
    0.64
     confirmed
    0.63
     truth
    0.62
    página
    0.61
    Fordítás
    0.61
    futbolista
    0.60
     statement
    0.57
    confirmed
    0.57
    Activations Density 0.507%

    No Known Activations