Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsBlogSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    3. GPT2-Small
    4. Attention Out - 128k
    5. 0-ATT_128K-OAI
    6. 13593
    Prev
    Next
    INDEX
    Explanations

    forms of past tense verbs and adjectives related to description or assessment

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    jbloom/GPT2-Small-OAI-v5-128k-attn-out-SAEs/v5_128k_layer_0
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    131,072
    Data Type
    torch.float32
    Hook Name
    blocks.0.hook_attn_out
    Hook Layer
    0
    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
    167
    +0.21
    1.6%
    99
    +0.17
    1.2%
    211
    +0.16
    1.2%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    664
    +0.21
    0.02
    536
    +0.17
    0.01
    544
    +0.16
    0.01
    Negative Logits
    nces
    -1.00
    OPLE
    -0.89
    occup
    -0.88
    Downloadha
    -0.88
    netflix
    -0.84
    qqa
    -0.83
    cius
    -0.81
    ibaba
    -0.80
    ories
    -0.78
    emia
    -0.77
    POSITIVE LOGITS
     corrid
    1.10
    ©¶æ
    0.95
    Ĥİ
    0.83
    athing
    0.80
    enance
    0.79
     defe
    0.78
    \\\\\\\\
    0.77
     misplaced
    0.75
    sburg
    0.69
    ©¶æ¥µ
    0.69
    Activations Density 0.023%

    No Known Activations