© 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. 1-RES-JB
    11. 16366
    Prev
    Next
    INDEX
    Explanations

    names of exclusive releases or updates

    oai_token-act-pair · gpt-3.5-turbo

    instances of the abbreviation "EX" in various contexts or entries

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    kowski
    -0.78
     Kuh
    -0.77
    kamp
    -0.72
    gaard
    -0.70
     lawn
    -0.70
     Bolton
    -0.69
     Wald
    -0.68
    yard
    -0.67
    arity
    -0.65
     Sut
    -0.64
    POSITIVE LOGITS
    CLUS
    1.24
    CEPT
    1.19
    TERN
    1.13
    PORT
    1.12
    EX
    1.05
    ACT
    1.05
    PER
    1.03
    ercise
    1.01
    PRESS
    1.01
    ERC
    1.01
    Activations Density 0.006%

    No Known Activations