© 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
    2. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    3. GPT2-Small
    4. Residual Post - 32k
    5. 8-RES_POST_32K-OAI
    6. 4703
    Prev
    Next
    INDEX
    Explanations

    instances of the word "Rec" and its variations, indicating a focus on recognition or recalling information

    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
    746
    +0.12
    0.6%
    176
    +0.11
    0.5%
    545
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    746
    +0.12
    0.03
    503
    +0.11
    0.03
    251
    +0.11
    0.04
    Negative Logits
    igans
    -0.77
    edIn
    -0.68
     kindred
    -0.64
    naires
    -0.60
    Reviewer
    -0.60
    ICAN
    -0.59
     Siberian
    -0.58
    hyde
    -0.57
    mates
    -0.55
    ãĥĹ
    -0.55
    POSITIVE LOGITS
    orpor
    1.03
    overed
    1.03
    ommod
    1.01
    urrent
    1.00
    rep
    0.99
    ursor
    0.97
    urrence
    0.96
    redited
    0.93
    redits
    0.93
    redit
    0.93
    Activations Density 0.073%

    No Known Activations