© 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. 4767
    Prev
    Next
    INDEX
    Explanations

    gerunds and adjectives related to life-altering changes or critical situations

    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
    408
    +0.11
    0.5%
    373
    +0.11
    0.5%
    94
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    408
    +0.11
    0.07
    362
    +0.11
    0.05
    210
    +0.10
    0.04
    Negative Logits
    odox
    -0.78
    OSP
    -0.76
    worn
    -0.76
    wang
    -0.73
    arnaev
    -0.69
    docs
    -0.68
    tiny
    -0.67
    then
    -0.66
    BIT
    -0.66
    OVER
    -0.65
    POSITIVE LOGITS
    âĢİ
    0.71
     epilepsy
    0.64
     ordeal
    0.63
    assadors
    0.60
     plunge
    0.59
     undertaking
    0.59
     seizures
    0.59
     trek
    0.59
     gamble
    0.58
     omn
    0.57
    Activations Density 0.421%

    No Known Activations