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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 15030
    Prev
    Next
    INDEX
    Explanations

    the presence of specific words related to violent events and actions

    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
    1385
    +0.13
    0.4%
    946
    +0.12
    0.3%
    906
    +0.10
    0.3%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    946
    +0.13
    0.06
    857
    +0.12
    0.03
    1352
    +0.10
    0.05
    Negative Logits
     soto
    -0.63
     plis
    -0.63
     ekos
    -0.60
     muna
    -0.59
     encre
    -0.59
     stopp
    -0.57
     cabrio
    -0.57
     habang
    -0.56
     Italijani
    -0.56
     pecuni
    -0.56
    POSITIVE LOGITS
     écout
    0.76
     fameux
    0.65
     découv
    0.61
     évit
    0.61
     curieux
    0.60
     parlant
    0.59
     conçus
    0.59
     offrant
    0.58
     réal
    0.58
     rassemb
    0.58
    Activations Density 0.337%

    No Known Activations