© 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. 7947
    Prev
    Next
    INDEX
    Explanations

    expressions related to achieving success or winning

    oai_token-act-pair · gpt-3.5-turbo

    references to achievements and victories in competitions or awards

    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
    zik
    -0.71
    iste
    -0.69
    ffer
    -0.66
     lapse
    -0.65
    APH
    -0.63
     arriving
    -0.62
    wards
    -0.62
    uders
    -0.62
    odder
    -0.60
    EEP
    -0.60
    POSITIVE LOGITS
     lottery
    1.01
     Pulitzer
    1.00
     contest
    0.96
     battles
    0.93
     championships
    0.90
     Nobel
    0.89
     contests
    0.89
     trophies
    0.89
     battle
    0.87
     prize
    0.87
    Activations Density 0.126%

    No Known Activations