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. 15549
    Prev
    Next
    INDEX
    Explanations

    mentions of people calling someone 'Gidget' and the resulting confusion

    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
    1967
    +0.27
    0.9%
    845
    +0.16
    0.5%
    674
    +0.15
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    845
    +0.27
    0.07
    1967
    +0.16
    0.06
    1801
    +0.15
    0.03
    Negative Logits
    Personensuche
    -0.57
    aarrggbb
    -0.55
    Xar
    -0.53
    RTGC
    -0.49
     Himo
    -0.48
    PCell
    -0.47
    Pautan
    -0.47
    pushFollow
    -0.47
    Hentet
    -0.46
     […]
    -0.46
    POSITIVE LOGITS
     shenan
    1.32
     unspeak
    1.23
     intersper
    1.22
     viciss
    1.22
     disreg
    1.20
     indestru
    1.17
     indescri
    1.16
     unve
    1.14
     encomp
    1.13
     maneu
    1.12
    Activations Density 1.322%

    No Known Activations