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. 8344
    Prev
    Next
    INDEX
    Explanations

    references to news articles or reports with a focus on various topics such as technology, politics, and crime

    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
    674
    +0.44
    7.3%
    1741
    +0.07
    1.1%
    1120
    +0.04
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    674
    +0.44
    0.93
    823
    +0.07
    0.93
    654
    +0.04
    0.94
    Negative Logits
    ে
    -1.12
     भी
    -1.11
     не
    -1.10
     ко
    -1.08
     про
    -1.08
    া
    -1.08
     но
    -1.08
    所有
    -1.08
    ার
    -1.07
     so
    -1.07
    POSITIVE LOGITS
    <bos>
    13.15
     encomp
    4.66
     fuf
    4.64
     guarante
    4.59
     affor
    4.58
     effe
    4.56
     squa
    4.55
     fta
    4.53
     increa
    4.52
     secon
    4.48
    Activations Density 0.869%

    No Known Activations