© 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. 4808
    Prev
    Next
    INDEX
    Explanations

    terms related to military and rescue operations

    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
    749
    +0.13
    0.6%
    205
    +0.11
    0.5%
    383
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    288
    +0.13
    0.07
    247
    +0.11
    0.05
    540
    +0.10
    0.06
    Negative Logits
     Cind
    -0.64
     Baptist
    -0.62
    ãģį
    -0.61
     Converted
    -0.61
    payment
    -0.61
    ãģĦ
    -0.61
     Flo
    -0.61
     Column
    -0.60
    imm
    -0.59
    owed
    -0.59
    POSITIVE LOGITS
     underway
    1.35
     undertaken
    1.26
     spearheaded
    1.09
     aimed
    1.08
     begun
    1.07
     waged
    1.06
     involving
    1.02
     conducted
    1.02
     initiated
    1.02
     abroad
    1.00
    Activations Density 0.416%

    No Known Activations