© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 5082
    Prev
    Next
    INDEX
    Explanations

    expressions of fear or concern

    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
    373
    +0.12
    0.5%
    483
    +0.11
    0.5%
    407
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    407
    +0.12
    0.03
    602
    +0.11
    0.02
    105
    +0.10
    0.03
    Negative Logits
    çĦ
    -0.79
    brates
    -0.72
    pta
    -0.71
    eness
    -0.71
    slice
    -0.69
    cence
    -0.68
    nice
    -0.68
     Kinnikuman
    -0.67
    ety
    -0.66
    OVA
    -0.66
    POSITIVE LOGITS
     retribution
    1.11
     repercussions
    1.01
     lest
    0.97
     repr
    0.88
     retaliation
    0.85
     pregn
    0.78
     imminent
    0.77
     impending
    0.71
     suicidal
    0.71
     unsafe
    0.69
    Activations Density 0.110%

    No Known Activations