© 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. 26108
    Prev
    Next
    INDEX
    Explanations

    references to the concepts of value, measurement, and expectation in various contexts

    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
    358
    +0.17
    1.0%
    23
    +0.12
    0.6%
    327
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    358
    +0.17
    0.13
    407
    +0.12
    0.12
    327
    +0.11
    0.13
    Negative Logits
    ¼
    -3.19
    İ
    -3.12
    ģ
    -3.09
    Ĵ
    -3.08
    Ĭ
    -3.08
    ¾
    -3.06
    µ
    -3.04
    Ģ
    -3.00
    ¢
    -2.96
    ¬
    -2.92
    POSITIVE LOGITS
    alities
    1.74
    exports
    1.67
    ostat
    1.48
     !==
    1.46
    oe
    1.46
    ôle
    1.46
    requently
    1.40
    ighting
    1.40
    ogn
    1.39
    ality
    1.38
    Activations Density 3.646%

    No Known Activations