Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Initial Gemma-2B-IT SAEs
    3. Gemma-2B-IT
    4. Residual Stream
    5. 12-RES-JB
    6. 2708
    Prev
    Next
    INDEX
    Explanations

    messages or communication-related content, particularly with a critical or informative tone

    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
    1839
    +0.16
    0.6%
    805
    +0.15
    0.6%
    331
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    805
    +0.16
    0.03
    1839
    +0.15
    0.03
    130
    +0.12
    0.03
    Negative Logits
     intit
    -0.69
     Luglio
    -0.64
    artney
    -0.63
     susun
    -0.62
     renou
    -0.60
     Giugno
    -0.59
     pitié
    -0.59
     migli
    -0.58
     apparti
    -0.58
     disegni
    -0.56
    POSITIVE LOGITS
     message
    1.37
    message
    1.23
     messages
    1.21
     Message
    1.19
     MESSAGE
    1.05
    Message
    1.04
     Messages
    1.03
     msg
    1.03
    messages
    1.00
     messaging
    0.98
    Activations Density 0.073%

    No Known Activations