Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Initial Gemma-2B-IT SAEs
    3. Gemma-2B-IT
    4. Residual Stream
    5. 12-RES-JB
    6. 8000
    Prev
    Next
    INDEX
    Explanations

    phrases related to physical violence

    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
    156
    +0.10
    0.3%
    946
    +0.08
    0.2%
    658
    +0.08
    0.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    466
    +0.10
    0.04
    946
    +0.08
    0.04
    822
    +0.08
    0.03
    Negative Logits
     parteci
    -0.96
     sappi
    -0.94
     credere
    -0.88
     rispond
    -0.88
     doveva
    -0.86
     glan
    -0.86
     haup
    -0.86
     gius
    -0.85
     gioca
    -0.84
     archi
    -0.84
    POSITIVE LOGITS
     beaten
    0.62
     beating
    0.60
     beat
    0.59
     beats
    0.56
     violently
    0.56
    beaten
    0.52
     violent
    0.52
     knocked
    0.49
     repeatedly
    0.48
     unconscious
    0.48
    Activations Density 0.214%

    No Known Activations