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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 13843
    Prev
    Next
    INDEX
    Explanations

    descriptions of military personnel and events related to military service

    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
    509
    +0.16
    0.5%
    658
    +0.09
    0.3%
    163
    +0.08
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    509
    +0.16
    0.10
    658
    +0.09
    0.08
    1038
    +0.08
    0.04
    Negative Logits
     imposs
    -0.64
     territo
    -0.63
     chay
    -0.63
     kasa
    -0.62
     faw
    -0.61
     thun
    -0.60
     seiz
    -0.60
     lidl
    -0.60
     mauri
    -0.60
     autob
    -0.58
    POSITIVE LOGITS
     training
    0.53
     officers
    0.52
     htons
    0.51
     officer
    0.51
     commander
    0.49
     military
    0.47
     Tarragona
    0.47
     camaraderie
    0.47
     PTSD
    0.46
     command
    0.45
    Activations Density 1.027%

    No Known Activations