© 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. 3534
    Prev
    Next
    INDEX
    Explanations

    concepts related to strategy and conflict

    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
    420
    +0.20
    0.9%
    732
    +0.12
    0.6%
    481
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    420
    +0.20
    -0.01
    732
    +0.12
    0.14
    237
    +0.10
    0.06
    Negative Logits
    kef
    -0.90
    ãĤ´ãĥ³
    -0.77
     Oath
    -0.74
    rompt
    -0.71
     documentaries
    -0.70
    ombat
    -0.70
     anniversary
    -0.68
     lol
    -0.68
    ocre
    -0.67
    hedon
    -0.67
    POSITIVE LOGITS
    GROUP
    0.75
    Chart
    0.70
    ++++++++++++++++
    0.69
     surplus
    0.69
    incre
    0.67
     longitudinal
    0.66
     respondent
    0.66
     Defendants
    0.66
     contiguous
    0.64
     âĪĴ
    0.64
    Activations Density 12.399%

    No Known Activations