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

    Neuronpedia

    APISteerSAE EvalsBlog/PodcastNEWSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlog/PodcastGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Initial Gemma-2B SAEs
    3. Gemma-2B
    4. Residual Stream
    5. 12-RES-JB
    6. 125
    Prev
    Next
    INDEX
    Explanations

    mentions of power outages and their impact on communities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    jbloom/Gemma-2b-Residual-Stream-SAEs/gemma_2b_blocks.12.hook_resid_post_16384
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    HuggingFaceFW/fineweb
    Features
    16,384
    Data Type
    torch.float32
    Hook Point
    blocks.12.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    HuggingFaceFW/fineweb
    Hook Point Layer
    12
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    50
    +0.32
    1.1%
    1120
    +0.12
    0.4%
    1842
    +0.10
    0.3%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1120
    +0.32
    0.01
    939
    +0.12
    0.07
    1385
    +0.10
    0.06
    Negative Logits
    <bos>
    -1.65
     germinate
    -0.54
     disbur
    -0.52
    Související
    -0.51
     synthesize
    -0.51
    Fordítás
    -0.51
     endow
    -0.50
     personalise
    -0.49
    deinit
    -0.48
    //---
    -0.47
    POSITIVE LOGITS
     bandung
    1.06
     Muhamma
    1.01
     jaya
    0.99
     lele
    0.89
     bangkok
    0.88
     reuters
    0.87
     uniqlo
    0.87
     Palembang
    0.87
     Jasa
    0.86
     seoul
    0.86
    Activations Density 0.708%

    No Known Activations