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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 14099
    Prev
    Next
    INDEX
    Explanations

    references to modifications or adjustments made to something

    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
    878
    +0.13
    0.4%
    795
    +0.13
    0.4%
    1407
    +0.12
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    878
    +0.13
    0.03
    1142
    +0.13
    0.02
    795
    +0.12
    0.02
    Negative Logits
     Marta
    -0.51
    Marta
    -0.48
     Villar
    -0.47
     Mónica
    -0.46
     Lucía
    -0.45
     Leopoldo
    -0.44
    FetchType
    -0.43
     fechar
    -0.42
    Lance
    -0.42
    Læs
    -0.41
    POSITIVE LOGITS
     modifications
    0.92
     modification
    0.92
     Modification
    0.90
     Modif
    0.89
     modifying
    0.86
     modif
    0.85
     Modifications
    0.84
     modifies
    0.84
     jaya
    0.83
    modification
    0.82
    Activations Density 0.089%

    No Known Activations