Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. MLP Post
    5. 0-MLP-SM
    6. 27644
    Prev
    Next
    INDEX
    Explanations

    the specific date "May 10"

    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
    444
    +0.12
    0.7%
    388
    +0.12
    0.7%
    173
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    173
    +0.12
    0.02
    4
    +0.12
    0.02
    153
    +0.12
    0.02
    Negative Logits
     quo
    -1.78
     hers
    -1.59
    rian
    -1.58
     theaters
    -1.56
     tears
    -1.56
     theirs
    -1.42
    rent
    -1.41
     damn
    -1.38
     kind
    -1.37
     true
    -1.37
    POSITIVE LOGITS
    bourne
    1.85
    flower
    1.80
    nals
    1.71
    iej
    1.67
    ning
    1.64
    aby
    1.58
    iour
    1.57
    avel
    1.55
    aceae
    1.54
    aceut
    1.52
    Activations Density 3.561%

    No Known Activations