© 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. Residual Post - 32k
    9. 8-RES_POST_32K-OAI
    11. 4799
    Prev
    Next
    INDEX
    Explanations

    conversational phrases that express opinions or questions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_8.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.8.hook_resid_post
    Hook Layer
    8
    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
    450
    +0.12
    0.5%
    727
    +0.11
    0.5%
    732
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    119
    +0.12
    0.05
    8
    +0.11
    0.06
    671
    +0.11
    0.06
    Negative Logits
    angering
    -0.72
    Interstitial
    -0.63
    20439
    -0.61
    âĸĪâĸĪâĸĪâĸĪâĸĪâĸĪâĸĪâĸĪ
    -0.60
    ©¶æ¥µ
    -0.59
     Excellence
    -0.58
    ifle
    -0.57
     otherwise
    -0.56
    Uncommon
    -0.56
    alions
    -0.55
    POSITIVE LOGITS
     yeah
    0.82
    bye
    0.76
     congratulations
    0.73
     dilemma
    0.71
    oiler
    0.70
     guess
    0.70
     wondering
    0.70
     begs
    0.69
     nutshell
    0.69
     huh
    0.69
    Activations Density 2.141%

    No Known Activations