© 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. 4504
    Prev
    Next
    INDEX
    Explanations

    positive adjectives and expressions of enjoyment

    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
    1013
    +0.15
    0.5%
    674
    +0.12
    0.3%
    184
    +0.09
    0.3%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1013
    +0.15
    0.06
    2031
    +0.12
    0.04
    1355
    +0.09
    0.04
    Negative Logits
     swarovski
    -0.92
     santiago
    -0.82
     oleo
    -0.81
     verona
    -0.80
     Ename
    -0.77
     lorenzo
    -0.76
     Intere
    -0.76
     ricardo
    -0.76
     murano
    -0.75
     ecru
    -0.75
    POSITIVE LOGITS
    <bos>
    0.66
     spesifik
    0.57
     ķ
    0.57
     yesterday
    0.57
     kompati
    0.55
     vokal
    0.54
     تضيفلها
    0.54
    AndEndTag
    0.54
    ."\
    0.53
     izvē
    0.53
    Activations Density 0.501%

    No Known Activations