© 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. 26491
    Prev
    Next
    INDEX
    Explanations

    prepositions, particularly "in" and "at," indicating locations or positions

    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
    176
    +0.14
    0.8%
    170
    +0.13
    0.7%
    146
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    170
    +0.14
    0.05
    14
    +0.13
    0.06
    466
    +0.12
    0.05
    Negative Logits
    kers
    -1.66
    ker
    -1.56
    STRA
    -1.49
     veterinary
    -1.40
    ly
    -1.40
    idine
    -1.37
    HM
    -1.33
    helial
    -1.31
    ,[@
    -1.28
    liest
    -1.28
    POSITIVE LOGITS
    ĨĴ
    2.42
    º
    2.37
    »¿
    2.35
    IJ
    2.26
    ĭ
    2.16
    ¼
    2.16
    £
    2.15
    ↵
    2.15
    č↵
    2.15
    č↵
    2.15
    Activations Density 0.339%

    No Known Activations