© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Initial Gemma-2B SAEs
    5. Gemma-2B
    7. Residual Stream
    9. 10-RES-JB
    11. 4862
    Prev
    Next
    INDEX
    Explanations

    instances of the substring "ha"

    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
    Neuron Alignment
    Index
    Value
    % of L₁
    50
    +0.23
    1.3%
    1145
    +0.13
    0.7%
    25
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1343
    +0.23
    0.08
    227
    +0.13
    0.07
    1145
    +0.13
    0.04
    Negative Logits
    <bos>
    -2.21
    ⓧ
    -0.72
    Δείτε
    -0.64
    ാൻ
    -0.61
    <?
    -0.60
    Εκ
    -0.58
     cooperate
    -0.58
     cooperated
    -0.58
    s
    -0.56
     relax
    -0.56
    POSITIVE LOGITS
     Infir
    1.11
     volon
    1.05
     véhic
    1.05
     citoy
    1.01
     Ename
    1.01
     Manufact
    1.00
     justement
    1.00
     quoique
    0.99
     vété
    0.99
    Prede
    0.98
    Activations Density 0.340%

    No Known Activations