© 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. 27125
    Prev
    Next
    INDEX
    Explanations

    instances of the word "separate."

    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
    169
    +0.13
    0.7%
    402
    +0.12
    0.7%
    480
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    179
    +0.13
    0.03
    239
    +0.12
    0.03
    356
    +0.11
    0.02
    Negative Logits
     ourselves
    -1.80
    ister
    -1.69
    evol
    -1.63
     herself
    -1.52
     emerging
    -1.50
     expressed
    -1.49
     evolving
    -1.48
    chy
    -1.46
     acquaint
    -1.39
    teen
    -1.39
    POSITIVE LOGITS
    disambiguation
    1.66
    fiddle
    1.54
     comeback
    1.52
    )',
    1.47
    Simplify
    1.46
     coup
    1.44
    golang
    1.44
    FLAGS
    1.43
    CTX
    1.42
    ('#
    1.42
    Activations Density 0.272%

    No Known Activations