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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 8-RES-JB
    6. 20712
    Prev
    Next
    INDEX
    Explanations

    acronyms in parentheses

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     blot
    -0.77
     baffled
    -0.72
     impe
    -0.72
     depress
    -0.69
     drift
    -0.68
     butcher
    -0.68
     incons
    -0.67
     glands
    -0.66
     swoop
    -0.66
     pressing
    -0.65
    POSITIVE LOGITS
    aka
    1.29
    UGC
    1.25
    formerly
    1.15
    sic
    1.12
    SG
    1.08
    also
    1.08
    PDF
    1.07
    NAS
    1.06
    HUD
    1.04
    including
    1.03
    Activations Density 0.099%

    No Known Activations