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 SAEs
    3. Gemma-2B
    4. Residual Stream
    5. 0-RES-JB
    6. 95
    Prev
    Next
    INDEX
    Explanations

    web portals or online platforms offering a variety of content and services

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    jbloom/Gemma-2b-Residual-Stream-SAEs/gemma_2b_blocks.0.hook_resid_post_16384_anthropic
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    HuggingFaceFW/fineweb
    Features
    16,384
    Data Type
    torch.float32
    Hook Point
    blocks.0.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    HuggingFaceFW/fineweb
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    50
    +0.11
    0.6%
    1870
    +0.05
    0.3%
    1150
    +0.03
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    300
    +0.11
    0.07
    1403
    +0.05
    0.06
    1440
    +0.03
    0.06
    Negative Logits
    <bos>
    -1.82
    ⓧ
    -1.20
    
    
    -1.13
    /**
    -1.06
    <?
    -1.03
    /*
    -0.88
    <?
    
    -0.87
    /*++
    -0.73
     springfox
    -0.72
    /***
    
    -0.71
    POSITIVE LOGITS
     Juf
    1.43
     maneu
    1.40
     lele
    1.35
     roul
    1.33
     Minang
    1.33
     affor
    1.31
     sovere
    1.30
     meis
    1.29
     Augu
    1.28
     maroc
    1.28
    Activations Density 0.366%

    No Known Activations