© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    5. GPT2-Small
    7. Residual Post - 32k
    9. 10-RES_POST_32K-OAI
    11. 29347
    Prev
    Next
    INDEX
    Explanations

    phrases that express perception or subjective judgments

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_10.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.10.hook_resid_post
    Hook Layer
    10
    Architecture
    standard
    Context Size
    64
    Dataset
    Skylion007/openwebtext
    Activation Function
    topk
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    524
    +0.13
    0.6%
    178
    +0.10
    0.5%
    37
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    486
    +0.13
    0.04
    55
    +0.10
    0.04
    331
    +0.10
    0.03
    Negative Logits
    instead
    -0.74
    pez
    -0.71
    maybe
    -0.68
     untold
    -0.67
    unknown
    -0.67
     never
    -0.66
    itton
    -0.66
    éĹĺ
    -0.65
     rather
    -0.65
     unsus
    -0.65
    POSITIVE LOGITS
     anymore
    1.40
     nor
    1.11
     any
    1.01
     necessarily
    0.97
     anything
    0.96
     anywhere
    0.95
     remotely
    0.90
     anybody
    0.89
     whatsoever
    0.85
     bothered
    0.84
    Activations Density 0.171%

    No Known Activations