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

    Neuronpedia

    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. 4721
    Prev
    Next
    INDEX
    Explanations

    occurrences of the sound "ch" or its variations in different 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
    373
    +0.12
    0.5%
    21
    +0.11
    0.5%
    126
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    232
    +0.12
    0.04
    373
    +0.11
    -0.01
    701
    +0.11
    0.02
    Negative Logits
    ãĤ¨ãĥ«
    -0.68
     sanity
    -0.68
     churn
    -0.67
     clock
    -0.62
     CLS
    -0.60
    mson
    -0.60
     deadlines
    -0.59
     branching
    -0.58
    odge
    -0.57
     Mondays
    -0.57
    POSITIVE LOGITS
    neum
    0.81
    inian
    0.78
    adian
    0.76
    ospace
    0.74
    ter
    0.73
    oplan
    0.73
    Streamer
    0.72
    sch
    0.71
    illes
    0.70
    ococ
    0.70
    Activations Density 0.186%

    No Known Activations