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. 8437
    Prev
    Next
    INDEX
    Explanations

    mentions of legal ages and legal restrictions related to marriage and abortion

    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
    1978
    +0.12
    0.3%
    1842
    +0.11
    0.3%
    453
    +0.10
    0.3%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    776
    +0.12
    0.05
    453
    +0.11
    0.05
    1978
    +0.10
    0.04
    Negative Logits
     unspeak
    -1.45
     apprehen
    -1.32
     indescri
    -1.26
     gaily
    -1.22
     vainly
    -1.21
     intersper
    -1.20
     intrigu
    -1.16
     shenan
    -1.13
     tolerably
    -1.11
     unwarran
    -1.11
    POSITIVE LOGITS
     karton
    1.69
     silikon
    1.65
     lapto
    1.55
     alkoh
    1.53
     makro
    1.52
     kask
    1.52
     lampa
    1.50
     mikrofon
    1.49
     elek
    1.47
     torba
    1.43
    Activations Density 0.247%

    No Known Activations