© 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 EvalsExportsinterp-engineNEWAPI 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. 30185
    Prev
    Next
    INDEX
    Explanations

    brackets and their associated formatting

    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
    489
    +0.12
    0.7%
    478
    +0.12
    0.6%
    280
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    494
    +0.12
    0.04
    209
    +0.12
    0.02
    389
    +0.11
    0.02
    Negative Logits
     processing
    -1.53
     whom
    -1.51
     told
    -1.45
    :**
    -1.42
     state
    -1.37
    scape
    -1.37
     overnight
    -1.36
     operations
    -1.36
     said
    -1.35
     cast
    -1.30
    POSITIVE LOGITS
    unreadable
    2.08
    ^](#
    1.69
    ClCompile
    1.66
    \*](#
    1.62
    OID
    1.58
    osterone
    1.58
    isexual
    1.55
    quer
    1.54
    STRA
    1.46
    lycer
    1.43
    Activations Density 1.573%

    No Known Activations