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. 27245
    Prev
    Next
    INDEX
    Explanations

    forms of the verb "to call."

    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
    376
    +0.20
    1.1%
    180
    +0.14
    0.8%
    503
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    503
    +0.20
    0.01
    180
    +0.14
    0.01
    58
    +0.13
    0.01
    Negative Logits
    itian
    -1.51
     answering
    -1.46
    uen
    -1.46
     emerged
    -1.44
     reached
    -1.39
    uke
    -1.38
    >'
    -1.35
    sen
    -1.34
    ript
    -1.34
    *=
    -1.33
    POSITIVE LOGITS
    ĻĤ
    1.86
    hearted
    1.71
    stance
    1.62
    ory
    1.47
     Wisconsin
    1.44
     Indiana
    1.43
    ware
    1.42
     goodness
    1.40
     Rapids
    1.39
    ential
    1.38
    Activations Density 0.014%

    No Known Activations