© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Initial Gemma-2B-IT SAEs
    5. Gemma-2B-IT
    7. Residual Stream
    9. 12-RES-JB
    11. 789
    Prev
    Next
    INDEX
    Explanations

    verbs denoting action or transformation, possibly related to technology or processes

    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
    50
    +0.22
    0.8%
    764
    +0.17
    0.6%
    394
    +0.14
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    2006
    +0.22
    0.07
    2044
    +0.17
    0.09
    1013
    +0.14
    0.08
    Negative Logits
    <bos>
    -1.46
    xpress
    -0.68
     ▼
    -0.64
     monta
    -0.63
     Distrikt
    -0.63
     CWE
    -0.63
     Kont
    -0.61
    ▼
    -0.61
    wel
    -0.60
     OMIT
    -0.60
    POSITIVE LOGITS
     unwarran
    2.35
     unspeak
    2.26
     disagre
    2.10
     tolerably
    2.03
     gaily
    2.00
     unlaw
    1.95
     Shakspeare
    1.94
     ecru
    1.94
     indescri
    1.90
     apprehen
    1.90
    Activations Density 0.951%

    No Known Activations