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. 10-RES-JB
    6. 1113
    Prev
    Next
    INDEX
    Explanations

    contractions and possessives, particularly related to negation and ownership

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    jbloom/Gemma-2b-Residual-Stream-SAEs/gemma_2b_blocks.10.hook_resid_post_16384
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    chanind/openwebtext-gemma
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.10.hook_resid_post
    Hook Layer
    10
    Architecture
    standard
    Context Size
    1,024
    Dataset
    ctigges/openwebtext-gemma-1024-cl
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    604
    +0.15
    0.7%
    478
    +0.15
    0.6%
    1967
    +0.14
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    478
    +0.15
    0.10
    1224
    +0.15
    0.07
    1253
    +0.14
    0.06
    Negative Logits
    <bos>
    -2.79
     intersper
    -2.29
     hairc
    -1.90
     hentai
    -1.87
     milf
    -1.81
     swarovski
    -1.73
     encomp
    -1.73
     snoopy
    -1.73
     inconce
    -1.71
     impra
    -1.69
    POSITIVE LOGITS
     optik
    0.88
     demokra
    0.83
     kalori
    0.81
     seksi
    0.78
    ’
    0.77
    mathrm
    0.77
     terci
    0.76
     keramik
    0.75
     elektronik
    0.74
     Meksiko
    0.73
    Activations Density 0.414%

    No Known Activations