© 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. 10-RES-JB
    11. 12717
    Prev
    Next
    INDEX
    Explanations

    phrases related to construction and progress updates

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating the status or progress of projects and developments

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     sidx
    -0.80
    mort
    -0.65
     sums
    -0.64
     spokes
    -0.63
    holder
    -0.63
     representations
    -0.63
     depictions
    -0.61
     Maher
    -0.60
     heirs
    -0.60
     descriptions
    -0.60
    POSITIVE LOGITS
     underway
    1.16
     halted
    1.11
     performed
    1.07
     done
    1.07
     resumed
    1.06
     occur
    1.04
     occurring
    1.03
     conducted
    0.99
     commence
    0.99
     undertaken
    0.98
    Activations Density 0.370%

    No Known Activations