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. 6896
    Prev
    Next
    INDEX
    Explanations

    scientific and technical terms related to research articles or technical documentation

    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
    871
    +0.16
    0.9%
    1053
    +0.13
    0.7%
    899
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    227
    +0.16
    0.09
    871
    +0.13
    0.06
    1343
    +0.13
    0.09
    Negative Logits
    <bos>
    -2.12
    /*
    -0.66
    ⓧ
    -0.64
     للمعارف
    -0.64
    <tfoot>
    -0.59
    TintMode
    -0.59
    aarrggbb
    -0.58
    glDisable
    -0.55
     وتسجيلات
    -0.55
    GEBURTSDATUM
    -0.54
    POSITIVE LOGITS
     territo
    1.03
     saar
    0.96
     wien
    0.91
     maroc
    0.91
     liev
    0.89
     uhr
    0.89
     saad
    0.89
     naer
    0.87
     pank
    0.87
     alha
    0.87
    Activations Density 0.923%

    No Known Activations