© 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. 5274
    Prev
    Next
    INDEX
    Explanations

    instances of the word "introduce" and its variations

    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
    12
    +0.11
    0.5%
    339
    +0.10
    0.5%
    591
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    651
    +0.11
    0.03
    746
    +0.10
    0.02
    503
    +0.10
    0.02
    Negative Logits
     denomin
    -0.72
     Mankind
    -0.67
     ILCS
    -0.64
     polyg
    -0.64
     Welfare
    -0.60
    ingly
    -0.59
    ively
    -0.59
     miscarriage
    -0.59
     srfAttach
    -0.58
     retard
    -0.58
    POSITIVE LOGITS
    ctory
    1.13
    ctions
    1.06
    cest
    1.04
    ches
    1.03
    ched
    0.98
    cing
    0.93
    gment
    0.91
    ¹
    0.91
    ¬
    0.89
    vous
    0.89
    Activations Density 0.120%

    No Known Activations