© 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
    3. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    5. GPT2-Small
    7. Attention Out - 32k
    9. 3-ATT_32K-OAI
    11. 3824
    Prev
    Next
    INDEX
    Explanations

    phrases related to contradictions or challenges in arguments

    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
    501
    +0.10
    0.5%
    417
    +0.10
    0.5%
    286
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    501
    +0.10
    0.04
    447
    +0.10
    0.06
    621
    +0.10
    0.03
    Negative Logits
    artifacts
    -0.75
    ãĤ°
    -0.70
    Reloaded
    -0.68
    butt
    -0.67
    venient
    -0.67
     fitt
    -0.63
    hole
    -0.63
     XXX
    -0.63
    ulative
    -0.63
     blu
    -0.62
    POSITIVE LOGITS
    depending
    0.77
     millenn
    0.71
    emale
    0.67
     alike
    0.64
     Algeria
    0.62
    IRED
    0.62
    TPPStreamerBot
    0.59
    Sham
    0.59
     eras
    0.58
    ratch
    0.57
    Activations Density 1.173%

    No Known Activations