© 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. Attention Out - 32k
    9. 3-ATT_32K-OAI
    11. 3374
    Prev
    Next
    INDEX
    Explanations

    instances of the word "manipulate" and its derivatives, indicating a focus on the concept of manipulation in various contexts

    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
    497
    +0.13
    0.6%
    642
    +0.10
    0.5%
    333
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    698
    +0.13
    0.01
    200
    +0.10
    0.01
    67
    +0.09
    0.01
    Negative Logits
    ukemia
    -0.85
    edIn
    -0.79
    ufact
    -0.78
    rentices
    -0.77
    hower
    -0.76
    ailability
    -0.74
    uilt
    -0.71
    pection
    -0.70
     condol
    -0.70
    terday
    -0.69
    POSITIVE LOGITS
     levers
    0.76
     wedge
    0.70
     tone
    0.69
    heim
    0.67
     BCC
    0.67
     Flavoring
    0.65
     surfaces
    0.63
    oidal
    0.62
     Hof
    0.62
     slides
    0.62
    Activations Density 0.147%

    No Known Activations