© 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. 11-RES-JB
    11. 22496
    Prev
    Next
    INDEX
    Explanations

    phrases related to actions or preparations before proceeding with something

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating actions that should occur prior to another event

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    rather
    -0.75
    ccording
    -0.70
     millenn
    -0.66
    while
    -0.66
     misunderstood
    -0.64
    paralle
    -0.64
     neglected
    -0.63
     dodged
    -0.63
     overlooked
    -0.63
    ĸļ
    -0.62
    POSITIVE LOGITS
     anymore
    0.97
     any
    0.86
     anything
    0.83
     final
    0.72
     anyone
    0.71
     anybody
    0.70
    anything
    0.69
     officially
    0.69
     attRot
    0.68
     ANY
    0.68
    Activations Density 0.281%

    No Known Activations