© 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. 10-RES-JB
    11. 6329
    Prev
    Next
    INDEX
    Explanations

    specific information about deadlines and instructions

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating future actions or outcomes, particularly those using the word "will."

    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
     mastering
    -0.75
    ebus
    -0.71
     bashing
    -0.70
    thinking
    -0.70
     manipulating
    -0.69
     squeezing
    -0.68
     inventive
    -0.67
    building
    -0.65
     hating
    -0.65
     slaying
    -0.65
    POSITIVE LOGITS
     be
    1.30
     expire
    1.16
     appear
    1.12
     receive
    1.12
     remain
    1.12
     automatically
    1.11
     notify
    1.08
     incur
    1.07
     determine
    1.07
     suffice
    1.03
    Activations Density 0.108%

    No Known Activations