© 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. 7400
    Prev
    Next
    INDEX
    Explanations

    words related to legal proceedings and applications

    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
    1352
    +0.10
    0.3%
    1150
    +0.09
    0.3%
    82
    +0.08
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    1553
    +0.10
    0.04
    1593
    +0.09
    0.03
    1352
    +0.08
    0.04
    Negative Logits
     utop
    -0.65
     alfabe
    -0.61
     solidar
    -0.61
     ideolog
    -0.60
     Deum
    -0.58
     Ause
    -0.58
     RSSSF
    -0.57
     Exacts
    -0.56
     violon
    -0.56
     auctor
    -0.55
    POSITIVE LOGITS
     unspeak
    1.39
     apprehen
    1.30
     shenan
    1.26
     horrend
    1.21
     ineffec
    1.18
     gaily
    1.17
     intersper
    1.17
     impra
    1.17
     vainly
    1.15
     maneu
    1.09
    Activations Density 0.235%

    No Known Activations