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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 3595
    Prev
    Next
    INDEX
    Explanations

    instances of the word "square" and related terms 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
    71
    +0.14
    0.6%
    362
    +0.09
    0.4%
    598
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    526
    +0.14
    0.01
    734
    +0.09
    0.02
    587
    +0.09
    0.02
    Negative Logits
    serving
    -0.74
    ridges
    -0.71
     aback
    -0.70
     inspected
    -0.65
    markets
    -0.63
     overboard
    -0.63
     froze
    -0.61
    seek
    -0.60
    istan
    -0.59
    ogg
    -0.59
    POSITIVE LOGITS
     princ
    0.79
     criteria
    0.77
    Reason
    0.69
     motto
    0.66
    arte
    0.65
     ethic
    0.64
    ornings
    0.64
     reality
    0.64
    Phys
    0.64
     paradigm
    0.63
    Activations Density 0.224%

    No Known Activations