© 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. 1097
    Prev
    Next
    INDEX
    Explanations

    words related to the concept of maintaining something

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "maintain" and variations of it

    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
     Holiday
    -0.70
     Shopping
    -0.64
     starring
    -0.64
     Christmas
    -0.63
    ookie
    -0.63
     Fields
    -0.61
     Robots
    -0.59
     Moments
    -0.59
     Choice
    -0.59
     Birthday
    -0.58
    POSITIVE LOGITS
     maintain
    3.33
     maintained
    2.19
    aintain
    2.18
     maintaining
    1.98
     maintains
    1.98
     retain
    1.73
     sustain
    1.57
     preserve
    1.50
     maintenance
    1.43
     keep
    1.40
    Activations Density 0.017%

    No Known Activations