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. 1189
    Prev
    Next
    INDEX
    Explanations

    conditional phrases and additions in the context of discussion

    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
    50
    +0.25
    0.9%
    1381
    +0.08
    0.3%
    1512
    +0.08
    0.3%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1473
    +0.25
    0.03
    1512
    +0.08
    0.03
    1799
    +0.08
    0.03
    Negative Logits
    <bos>
    -2.66
    ⓧ
    -0.61
     disbur
    -0.59
     Roskov
    -0.55
     gynhyrchwyd
    -0.55
    dirig
    -0.55
    ContentAlignment
    -0.54
    Autoritní
    -0.53
     dirig
    -0.51
    Personendaten
    -0.51
    POSITIVE LOGITS
     vklju
    0.77
     benzina
    0.77
    favourable
    0.77
     saad
    0.76
     paradiso
    0.75
    solicited
    0.74
    mistak
    0.74
     Karang
    0.73
    gruntled
    0.72
     izdel
    0.71
    Activations Density 0.238%

    No Known Activations