© 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. 8-RES_POST_32K-OAI
    11. 5902
    Prev
    Next
    INDEX
    Explanations

    recurring punctuation, specifically commas

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_8.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.8.hook_resid_post
    Hook Layer
    8
    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
    501
    +0.12
    0.6%
    157
    +0.10
    0.5%
    581
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    460
    +0.12
    0.02
    43
    +0.10
    0.02
    370
    +0.10
    0.02
    Negative Logits
    mates
    -0.62
    sword
    -0.60
     Flavoring
    -0.60
    hari
    -0.60
    tnc
    -0.59
     highs
    -0.58
    ories
    -0.57
     Yourself
    -0.57
    SourceFile
    -0.56
    tops
    -0.56
    POSITIVE LOGITS
    âķIJâķIJ
    0.79
     unlike
    0.77
    adle
    0.77
    interstitial
    0.69
     anecd
    0.67
    requires
    0.67
    artifacts
    0.66
    acha
    0.66
    ocobo
    0.63
    âĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢ
    0.63
    Activations Density 0.161%

    No Known Activations