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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 27269
    Prev
    Next
    INDEX
    Explanations

    prepositions used in infinitive phrases

    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
    138
    +0.14
    0.7%
    180
    +0.13
    0.7%
    240
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    69
    +0.14
    0.38
    151
    +0.13
    0.31
    434
    +0.12
    0.30
    Negative Logits
    ·¸
    -1.59
    oses
    -1.48
    angling
    -1.48
     Wrote
    -1.48
    irection
    -1.48
    ados
    -1.46
    cuts
    -1.43
     recipients
    -1.36
    imore
    -1.34
    ŀ
    -1.33
    POSITIVE LOGITS
    rame
    1.44
    creen
    1.41
    ycler
    1.40
     Alexa
    1.37
    oud
    1.37
     morning
    1.36
    ged
    1.34
     chloride
    1.34
    ILY
    1.32
    defense
    1.29
    Activations Density 1.625%

    No Known Activations