© 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. 24568
    Prev
    Next
    INDEX
    Explanations

    sequences of brackets or brackets enclosed around content

    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
    98
    +0.13
    0.7%
    142
    +0.12
    0.7%
    450
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    56
    +0.13
    0.04
    68
    +0.12
    0.03
    461
    +0.11
    0.03
    Negative Logits
    ŀ
    -2.96
    Ĺ
    -2.79
    ¿½
    -2.52
    ij
    -2.47
    ·¸
    -2.46
    ¥
    -2.44
    ī
    -2.42
    Ĥ¬
    -2.40
    »¿
    -2.32
    ¡
    -2.31
    POSITIVE LOGITS
    swift
    1.70
    win
    1.68
    event
    1.61
    emb
    1.60
    ractor
    1.58
    vette
    1.58
    vier
    1.54
    emann
    1.53
     Exped
    1.44
    Jackson
    1.42
    Activations Density 0.167%

    No Known Activations