© 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
    2. Joseph Bloom · Initial Gemma-2B-IT SAEs
    3. Gemma-2B-IT
    4. Residual Stream
    5. 12-RES-JB
    6. 7783
    Prev
    Next
    INDEX
    Explanations

    words related to technical details, features, or improvements in products or systems

    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
    961
    +0.09
    0.3%
    1385
    +0.09
    0.2%
    1439
    +0.07
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    919
    +0.09
    0.03
    961
    +0.09
    0.04
    1662
    +0.07
    0.05
    Negative Logits
     accla
    -1.70
     emphat
    -1.69
     embra
    -1.66
     dises
    -1.62
     aen
    -1.62
     increa
    -1.61
     guarante
    -1.60
     fuf
    -1.58
     fta
    -1.58
     depic
    -1.57
    POSITIVE LOGITS
     by
    0.75
    ReusableCell
    0.72
    awtextra
    0.71
     significantly
    0.71
     into
    0.66
     considerably
    0.65
     due
    0.64
    INCREF
    0.64
     than
    0.64
     towards
    0.64
    Activations Density 0.362%

    No Known Activations