© 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. 11509
    Prev
    Next
    INDEX
    Explanations

    verbs related to searching or striving towards a goal

    oai_token-act-pair · gpt-3.5-turbo

    phrases that include the term "seeks" indicating intent or desire for something

    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
     tipping
    -0.59
     dumb
    -0.57
     Ding
    -0.57
     right
    -0.56
    EA
    -0.56
     belt
    -0.56
     alarm
    -0.55
     landing
    -0.55
     inside
    -0.54
     up
    -0.54
    POSITIVE LOGITS
     seeks
    3.19
     strives
    2.07
     seek
    1.93
     intends
    1.72
     proposes
    1.71
     sought
    1.68
     aims
    1.64
     explores
    1.64
     accuses
    1.60
    seek
    1.57
    Activations Density 0.017%

    No Known Activations