© 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. 4931
    Prev
    Next
    INDEX
    Explanations

    instances of the word "as" used in various 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
    655
    +0.10
    0.4%
    369
    +0.10
    0.4%
    724
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    647
    +0.10
    0.07
    172
    +0.10
    0.06
    192
    +0.09
    0.05
    Negative Logits
     Archbishop
    -0.64
     editor
    -0.63
    zman
    -0.61
     saint
    -0.61
     Chancellor
    -0.60
     unanimous
    -0.60
     CEO
    -0.59
     archetype
    -0.59
     publisher
    -0.58
     businessman
    -0.57
    POSITIVE LOGITS
     opposed
    0.91
    instead
    0.84
     thereby
    0.76
    onds
    0.74
    result
    0.74
    shown
    0.71
    è£ħ
    0.71
    onga
    0.71
     Whereas
    0.70
    icone
    0.68
    Activations Density 0.243%

    No Known Activations