© 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. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 0-RES-JB
    11. 6135
    Prev
    Next
    INDEX
    Explanations

    mentions of specific types of missions that involve tasks or objectives

    oai_token-act-pair · gpt-3.5-turbo

    references to various types of missions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    owler
    -0.67
     Bernstein
    -0.66
     flush
    -0.62
     bracket
    -0.61
    ç
    -0.61
    chair
    -0.60
    Ļ½
    -0.60
    Bir
    -0.60
     Taj
    -0.60
    zbek
    -0.59
    POSITIVE LOGITS
     missions
    3.74
     Missions
    2.88
     mission
    2.47
    Mission
    2.04
     Mission
    2.03
     quests
    1.87
     objectives
    1.43
     deployments
    1.38
     assignments
    1.36
     flights
    1.35
    Activations Density 0.015%

    No Known Activations