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. 8252
    Prev
    Next
    INDEX
    Explanations

    references to ancient history or civilizations

    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
    897
    +0.13
    0.4%
    421
    +0.11
    0.4%
    1222
    +0.11
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1443
    +0.13
    0.02
    58
    +0.11
    0.02
    1137
    +0.11
    0.03
    Negative Logits
    TagNumber
    -0.47
    WireFormat
    -0.46
    ┴
    -0.46
    ઊ
    -0.45
     jestli
    -0.44
     gynnwys
    -0.42
    houette
    -0.42
    ToTensor
    -0.42
    Mientras
    -0.42
    Dispon
    -0.42
    POSITIVE LOGITS
     ancient
    0.85
     Ancient
    0.83
    Ancient
    0.80
     Traité
    0.78
     Mère
    0.78
    ancient
    0.75
     basili
    0.74
     Valentín
    0.74
     Mémoires
    0.74
     Præ
    0.73
    Activations Density 0.118%

    No Known Activations