© 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
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. MLP Post
    9. 2-MLP-SM
    11. 26714
    Prev
    Next
    INDEX
    Explanations

    phrases indicating claims, allegations, or assertions about events or behaviors

    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
    335
    +0.12
    0.7%
    431
    +0.11
    0.6%
    175
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    335
    +0.12
    0.12
    188
    +0.11
    0.02
    223
    +0.11
    0.03
    Negative Logits
    ---|---
    -1.68
    GBT
    -1.68
    ipes
    -1.54
    iph
    -1.52
    ---|---|---
    -1.51
    solete
    -1.48
    endered
    -1.46
    rangle
    -1.43
    dagger
    -1.42
    odor
    -1.40
    POSITIVE LOGITS
    manship
    1.98
    ño
    1.97
     anyway
    1.72
    »¿
    1.72
    hart
    1.71
    giving
    1.71
    gard
    1.68
     yet
    1.66
    ĭ
    1.65
    ĻĤ
    1.64
    Activations Density 2.125%

    No Known Activations