© 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
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 10-RES-JB
    11. 7777
    Prev
    Next
    INDEX
    Explanations

    references to military tactics and weapons

    oai_token-act-pair · gpt-3.5-turbo

    elements related to combat strategy and tactics

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ploma
    -0.73
     offic
    -0.71
     Duff
    -0.71
     Pric
    -0.70
     Bleach
    -0.70
    Parents
    -0.70
    ISTORY
    -0.70
     Guinness
    -0.68
    soDeliveryDate
    -0.68
     parenting
    -0.68
    POSITIVE LOGITS
     enemy
    1.46
     enemies
    1.35
    enemy
    1.26
     foes
    1.24
     targets
    1.21
     melee
    1.17
     attackers
    1.17
     foe
    1.16
     infantry
    1.16
     ambush
    1.12
    Activations Density 0.366%

    No Known Activations