© 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. 10297
    Prev
    Next
    INDEX
    Explanations

    phrases related to technical details and specifications

    oai_token-act-pair · gpt-3.5-turbo

    references to film production and completion

    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
     dogs
    -0.63
    Merit
    -0.62
     perfect
    -0.61
     Peak
    -0.61
     goats
    -0.61
     flying
    -0.59
     epic
    -0.59
     poor
    -0.59
     heroic
    -0.59
     Sard
    -0.59
    POSITIVE LOGITS
    tion
    4.77
    tions
    3.75
     tion
    1.99
    tor
    1.64
    TION
    1.48
    tan
    1.26
    iton
    1.23
    teness
    1.23
    ties
    1.22
    tin
    1.21
    Activations Density 0.007%

    No Known Activations