© 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. 2631
    Prev
    Next
    INDEX
    Explanations

    phrases related to Parmesan cheese and throttle control

    oai_token-act-pair · gpt-3.5-turbo

    references to specific food items, particularly Parmesan cheese, as well as terms related to measuring and enhancing flavor

    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
    */(
    -0.79
    HCR
    -0.76
    ierrez
    -0.75
    coded
    -0.74
    IBLE
    -0.73
     GHC
    -0.69
    IFIED
    -0.63
    NEY
    -0.63
     Cornell
    -0.62
    arians
    -0.62
    POSITIVE LOGITS
     Mile
    1.25
    gram
    0.88
    oshenko
    0.86
     Seym
    0.86
    bum
    0.74
    tti
    0.73
    ret
    0.71
    ujah
    0.71
     Duc
    0.69
     Thro
    0.69
    Activations Density 0.014%

    No Known Activations