© 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. 4872
    Prev
    Next
    INDEX
    Explanations

    references to statistical data and analysis related to various topics

    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
    229
    +0.10
    0.4%
    373
    +0.10
    0.4%
    122
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    680
    +0.10
    0.05
    362
    +0.10
    0.08
    625
    +0.10
    0.04
    Negative Logits
    Appearance
    -0.74
     detachment
    -0.67
    sburg
    -0.67
     indifferent
    -0.65
     murdering
    -0.64
    Interstitial
    -0.63
    opath
    -0.63
     isolation
    -0.62
    osexual
    -0.62
     inferior
    -0.61
    POSITIVE LOGITS
     Bloomberg
    0.97
     Nielsen
    0.93
     Eater
    0.90
     Thomson
    0.85
     Deadline
    0.84
     Freedom
    0.82
     POLITICO
    0.81
     Gallup
    0.78
     FOIA
    0.76
     census
    0.76
    Activations Density 1.306%

    No Known Activations