Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APISteerSAE EvalsBlog/PodcastNEWSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlog/PodcastGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 3-RES-JB
    6. 22061
    Prev
    Next
    INDEX
    Explanations

    words related to the composer "Beethoven"

    oai_token-act-pair · gpt-3.5-turbo

    references to the chemical compound "eth."

    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
     Clement
    -0.73
    è¦ļéĨĴ
    -0.70
     referen
    -0.66
    enhagen
    -0.66
    éĹĺ
    -0.64
    ļéĨĴ
    -0.60
    vier
    -0.60
     ital
    -0.60
     plaster
    -0.59
     releg
    -0.58
    POSITIVE LOGITS
    lehem
    1.33
    amphetamine
    1.14
    inks
    1.00
    iop
    0.95
    icone
    0.91
    ought
    0.91
    anol
    0.90
    ylene
    0.88
    ulhu
    0.87
    rones
    0.84
    Activations Density 0.008%

    No Known Activations