© 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. 10-RES_POST_32K-OAI
    6. 29581
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "tim" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_10.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.10.hook_resid_post
    Hook Layer
    10
    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
    536
    +0.12
    0.5%
    184
    +0.10
    0.4%
    619
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    314
    +0.12
    -0.00
    232
    +0.10
    0.02
    503
    +0.09
    0.01
    Negative Logits
    Ĥİ
    -0.71
     Pwr
    -0.69
     reservoirs
    -0.67
    onne
    -0.61
     Crown
    -0.56
     hon
    -0.55
    loo
    -0.55
     Sau
    -0.54
     neighb
    -0.54
     Union
    -0.53
    POSITIVE LOGITS
    eline
    1.15
    elist
    1.15
    bered
    1.14
    etr
    1.12
    idity
    1.11
    ework
    1.10
    encer
    1.10
    emen
    1.06
    eways
    1.04
    eless
    1.04
    Activations Density 0.135%

    No Known Activations