© 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. 3-RES-JB
    11. 9162
    Prev
    Next
    INDEX
    Explanations

    information related to the physical release of products

    oai_token-act-pair · gpt-3.5-turbo

    references to physical and digital versions of games

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ilus
    -0.78
     scapego
    -0.77
     slime
    -0.73
     probing
    -0.73
     thorn
    -0.69
    tackle
    -0.67
     aspirin
    -0.67
     shrinking
    -0.66
     dictatorship
    -0.65
     Tyrann
    -0.65
    POSITIVE LOGITS
    Tickets
    1.01
     TBA
    0.93
     Finals
    0.92
     Celebration
    0.89
    iversary
    0.89
    GMT
    0.88
     Anniversary
    0.87
    Xbox
    0.87
    cember
    0.85
     finals
    0.85
    Activations Density 0.815%

    No Known Activations