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. 11-RES-JB
    6. 15470
    Prev
    Next
    INDEX
    Explanations

    multiple occurrences of the same word

    oai_token-act-pair · gpt-3.5-turbo

    phrase patterns emphasizing quantities or groups

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Rhodes
    -0.70
     toast
    -0.60
     Newark
    -0.60
     300
    -0.59
     Hercules
    -0.57
     Brighton
    -0.56
     darts
    -0.54
     brill
    -0.53
     2100
    -0.53
     Jericho
    -0.53
    POSITIVE LOGITS
    âĢ
    1.91
     âĢ
    1.54
    ãĢ
    1.38
    âĶ
    1.26
    âĿ
    1.22
    Æ
    1.18
    âĢł
    1.17
    âĸ
    1.15
    ¨
    1.14
     âĶĤ
    1.14
    Activations Density 0.471%

    No Known Activations