© 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. 4110
    Prev
    Next
    INDEX
    Explanations

    concepts related to pedestrian safety and traffic management

    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
    720
    +0.12
    0.5%
    762
    +0.10
    0.5%
    598
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    225
    +0.12
    0.10
    260
    +0.10
    0.09
    570
    +0.10
    0.09
    Negative Logits
    MQ
    -0.80
     Patreon
    -0.79
    DragonMagazine
    -0.78
     Alchemist
    -0.77
    angel
    -0.73
    Dat
    -0.73
     bombshell
    -0.73
     notebooks
    -0.72
    canon
    -0.72
     rumours
    -0.71
    POSITIVE LOGITS
     congest
    0.90
    angering
    0.87
     obstruct
    0.86
     practicable
    0.85
     pedestrian
    0.84
     outdoors
    0.82
     towed
    0.79
     airborne
    0.79
     effected
    0.78
     prone
    0.76
    Activations Density 6.158%

    No Known Activations