© 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. 2-MLP-SM
    11. 26575
    Prev
    Next
    INDEX
    Explanations

    quantitative comparisons and expressions of magnitude

    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
    23
    +0.18
    1.0%
    497
    +0.18
    1.0%
    152
    +0.11
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    497
    +0.18
    0.00
    440
    +0.18
    0.00
    462
    +0.11
    0.01
    Negative Logits
    anks
    -2.05
    agna
    -1.72
    İ
    -1.48
    Į
    -1.48
    eries
    -1.48
    itudes
    -1.47
    Īĺ
    -1.46
     existed
    -1.45
    bara
    -1.40
    ough
    -1.38
    POSITIVE LOGITS
     [[*
    1.66
     suspicion
    1.55
    &\
    1.50
     formaldehyde
    1.48
    :&
    1.47
    &=\
    1.47
     sprinkle
    1.43
     nod
    1.42
     complaining
    1.41
     scatter
    1.40
    Activations Density 0.862%

    No Known Activations