© 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. 9-RES-JB
    11. 15757
    Prev
    Next
    INDEX
    Explanations

    phrases related to actions or intentions involving going, doing, or saying

    oai_token-act-pair · gpt-3.5-turbo

    expressions of intention or desire related to actions and activities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 9-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    9
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    requires
    -0.75
     Increasing
    -0.69
    marked
    -0.68
     unsurprisingly
    -0.67
     flagged
    -0.66
    surprisingly
    -0.65
     millenn
    -0.65
     Advantage
    -0.65
     strikingly
    -0.64
     noteworthy
    -0.63
    POSITIVE LOGITS
     stay
    1.32
     get
    1.18
     participate
    1.18
     finish
    1.13
     talk
    1.12
     listen
    1.10
     survive
    1.08
     hear
    1.07
     marry
    1.06
     communicate
    1.06
    Activations Density 0.414%

    No Known Activations