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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. MLP Post
    9. 2-MLP-SM
    11. 24454
    Prev
    Next
    INDEX
    Explanations

    references to Russia and its political landscape

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__mlp-sm_processed/2-mlp-sm
    Prompts (Dashboard)
    32,768 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.2.hook_mlp_out
    Hook Layer
    2
    Architecture
    standard
    Context Size
    128
    Dataset
    EleutherAI/the_pile_deduplicated
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    328
    +0.13
    0.7%
    315
    +0.13
    0.7%
    163
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    213
    +0.13
    0.04
    328
    +0.13
    0.03
    461
    +0.12
    0.03
    Negative Logits
    rus
    -1.70
    oths
    -1.49
    hest
    -1.46
    nt
    -1.45
    stood
    -1.45
    rous
    -1.45
    ht
    -1.42
    ward
    -1.41
    hton
    -1.40
     politicians
    -1.40
    POSITIVE LOGITS
    »¿
    2.72
    ¸
    2.52
    ¿
    2.32
    ľ
    2.30
    ł
    2.27
    ĭ
    2.27
    ¤
    2.23
    Ĭ
    2.21
    »
    2.13
    Ħ
    2.12
    Activations Density 0.439%

    No Known Activations