© 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
    2. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    3. GPT2-Small
    4. Attention Out - 32k
    5. 3-ATT_32K-OAI
    6. 3419
    Prev
    Next
    INDEX
    Explanations

    references to social interactions and community dynamics

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-att_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-attn-out-SAEs/v5_32k_layer_3
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.3.hook_attn_out
    Hook Layer
    3
    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
    745
    +0.14
    0.6%
    616
    +0.12
    0.5%
    481
    +0.12
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    516
    +0.14
    0.03
    256
    +0.12
    0.03
    616
    +0.12
    0.03
    Negative Logits
     Say
    -0.75
     Yak
    -0.74
    ORK
    -0.70
     Stay
    -0.70
     Mehran
    -0.69
     Din
    -0.69
     Zeit
    -0.66
     Mustang
    -0.65
     Esper
    -0.65
     Herz
    -0.65
    POSITIVE LOGITS
     failures
    0.77
    inventoryQuantity
    0.74
     whims
    0.74
     violations
    0.73
     constraints
    0.70
     failings
    0.70
     limitations
    0.69
     deficiencies
    0.68
     disruptions
    0.68
     disciplines
    0.67
    Activations Density 0.537%

    No Known Activations