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. 27431
    Prev
    Next
    INDEX
    Explanations

    the concept of transformation or change, particularly in relation to processes or paths

    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
    292
    +0.14
    0.8%
    61
    +0.13
    0.7%
    302
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    154
    +0.14
    0.06
    329
    +0.13
    0.06
    14
    +0.13
    0.06
    Negative Logits
     prospective
    -1.67
    ely
    -1.63
    agment
    -1.61
    oved
    -1.58
    ocene
    -1.51
     current
    -1.51
     %%
    -1.43
     elastic
    -1.42
    ceptible
    -1.39
    autiful
    -1.39
    POSITIVE LOGITS
    Bag
    1.74
    kie
    1.68
    ibles
    1.63
    iffs
    1.58
    night
    1.49
    ney
    1.45
    yards
    1.45
    ids
    1.41
    mans
    1.35
    drug
    1.33
    Activations Density 0.038%

    No Known Activations