© 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. 5330
    Prev
    Next
    INDEX
    Explanations

    instances of deception or manipulation within relationships and contexts

    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
    610
    +0.11
    0.5%
    139
    +0.09
    0.4%
    677
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    362
    +0.11
    0.06
    608
    +0.09
    0.06
    669
    +0.09
    0.04
    Negative Logits
    ————
    -0.76
    DragonMagazine
    -0.73
    =-=-=-=-
    -0.69
    è£ħ
    -0.69
    leigh
    -0.68
    ratulations
    -0.67
    ril
    -0.67
    Generic
    -0.65
    ========
    -0.65
    ensual
    -0.65
    POSITIVE LOGITS
     circulation
    0.91
     shelves
    0.83
     worn
    0.73
     existence
    0.72
     clut
    0.71
     rubble
    0.67
     jaws
    0.66
     landfill
    0.66
     cooldown
    0.66
    pload
    0.65
    Activations Density 0.335%

    No Known Activations