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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. MLP Post
    9. 0-MLP-SM
    11. 27228
    Prev
    Next
    INDEX
    Explanations

    expressions of optimism or potential success

    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/0-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.0.hook_mlp_out
    Hook Layer
    0
    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
    376
    +0.18
    1.0%
    156
    +0.13
    0.7%
    51
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    265
    +0.18
    0.01
    51
    +0.13
    0.01
    408
    +0.13
    0.01
    Negative Logits
    ello
    -1.84
    anson
    -1.60
    udi
    -1.59
    ardi
    -1.58
    ietz
    -1.55
    ing
    -1.54
    eller
    -1.46
    []{
    -1.43
    que
    -1.42
     dioxide
    -1.40
    POSITIVE LOGITS
    ł
    3.31
    ĸ
    3.15
    ¢
    3.12
    ¼
    2.99
    Ļ
    2.98
    ĥ
    2.96
    Ĥ¬
    2.96
    Īĺ
    2.91
    ļ
    2.91
    ¦
    2.82
    Activations Density 0.018%

    No Known Activations