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. 962
    Prev
    Next
    INDEX
    Explanations

    phrases indicating causation or reasons

    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
    +0.22
    1.3%
    1937
    +0.13
    0.7%
    680
    +0.09
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    765
    +0.22
    0.03
    370
    +0.13
    0.03
    1937
    +0.09
    0.03
    Negative Logits
    <bos>
    -3.19
     explore
    -0.67
    @
    -0.67
     hold
    -0.64
    cat
    -0.63
    放
    -0.62
     manage
    -0.62
     put
    -0.61
     integrate
    -0.60
     consult
    -0.59
    POSITIVE LOGITS
     maneu
    1.64
     increa
    1.60
     affor
    1.60
     embodi
    1.55
     unwarran
    1.52
     impra
    1.52
     ecru
    1.50
     reluct
    1.47
     Manufact
    1.47
     stockholm
    1.46
    Activations Density 0.033%

    No Known Activations