© 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. 0-RES-JB
    11. 7531
    Prev
    Next
    INDEX
    Explanations

    verbs related to saving, salvaging, or scavenging

    oai_token-act-pair · gpt-3.5-turbo

    terms related to recovery or rescue efforts, especially in the context of recovering items or preventing loss

    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
     ellipt
    -1.00
     monarch
    -0.80
     Tablet
    -0.74
    doms
    -0.70
    bishop
    -0.70
    20439
    -0.69
     persecuted
    -0.68
     stature
    -0.68
     Yen
    -0.65
    tex
    -0.64
    POSITIVE LOGITS
     salv
    2.33
     salvage
    2.24
     scaven
    2.01
     sabotage
    1.96
     sabot
    1.62
     improv
    1.47
     fireworks
    1.34
     Salv
    1.32
     sneak
    1.22
     Impro
    1.09
    Activations Density 0.059%

    No Known Activations