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 SAEs
    3. Gemma-2B
    4. Residual Stream
    5. 6-RES-JB
    6. 841
    Prev
    Next
    INDEX
    Explanations

    phrases related to religious symbolism and prophecies

    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.6.hook_resid_post_16384_anthropic_fast_lr
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    HuggingFaceFW/fineweb
    Features
    16,384
    Data Type
    torch.float32
    Hook Point
    blocks.6.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    HuggingFaceFW/fineweb
    Hook Point Layer
    6
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    50
    +0.29
    1.7%
    1741
    +0.21
    1.2%
    1967
    +0.16
    0.9%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1967
    +0.29
    0.07
    1581
    +0.21
    0.05
    1535
    +0.16
    0.04
    Negative Logits
    <bos>
    -2.71
    USTAIN
    -0.65
    contentLoaded
    -0.63
    ///**
    -0.63
    ğlık
    -0.60
    lış
    -0.59
    //};
    -0.59
     kasarigan
    -0.57
    otheby
    -0.57
    disciplinar
    -0.55
    POSITIVE LOGITS
     affor
    0.94
     accla
    0.91
     opport
    0.90
     emphat
    0.90
     Keny
    0.87
     maneu
    0.86
     valencia
    0.84
     Jambi
    0.84
     increa
    0.84
     inev
    0.83
    Activations Density 0.693%

    No Known Activations