© 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. 6226
    Prev
    Next
    INDEX
    Explanations

    phrases related to things remaining in a certain state or location over time

    oai_token-act-pair · gpt-3.5-turbo

    repeated phrases associated with consistency or maintenance

    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
    psey
    -0.64
     suspect
    -0.56
     complete
    -0.56
     estimate
    -0.54
     hope
    -0.53
     call
    -0.53
    ¿
    -0.53
     mean
    -0.51
     sound
    -0.51
     qualified
    -0.51
    POSITIVE LOGITS
     stays
    2.18
     grows
    1.92
     loses
    1.91
     waits
    1.84
     develops
    1.83
     listens
    1.82
     rises
    1.80
     lasts
    1.78
     evolves
    1.77
     fades
    1.74
    Activations Density 0.057%

    No Known Activations