© 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
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. MLP Post
    5. 2-MLP-SM
    6. 26802
    Prev
    Next
    INDEX
    Explanations

    emotional expressions and discussions about feelings

    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
    0.8%
    425
    +0.12
    0.7%
    114
    +0.11
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    364
    +0.13
    0.07
    98
    +0.12
    0.09
    456
    +0.11
    0.00
    Negative Logits
    leading
    -1.70
    sects
    -1.67
     healing
    -1.42
    ![**
    -1.39
     creation
    -1.34
    rbrace
    -1.32
    orph
    -1.30
    Inflater
    -1.28
    sect
    -1.27
    posing
    -1.26
    POSITIVE LOGITS
    ¸
    1.69
     год
    1.49
    ppa
    1.46
    ney
    1.42
    İ
    1.32
    sten
    1.31
    ollo
    1.30
    ¶
    1.29
     PUBL
    1.27
    hui
    1.25
    Activations Density 2.728%

    No Known Activations