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. 0-RES-JB
    6. 24491
    Prev
    Next
    INDEX
    Explanations

    sentences where someone expresses information to another person

    oai_token-act-pair · gpt-3.5-turbo

    instances of the verb "told."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     circulation
    -0.62
    izable
    -0.59
    Category
    -0.59
     elimination
    -0.58
    crew
    -0.57
    animate
    -0.56
     constit
    -0.55
     upward
    -0.55
     veh
    -0.55
     scratch
    -0.54
    POSITIVE LOGITS
     told
    3.26
    told
    2.39
     tells
    2.02
     warned
    1.72
     explained
    1.66
     instructed
    1.66
     asked
    1.64
     recounted
    1.62
     texted
    1.61
     testified
    1.59
    Activations Density 0.059%

    No Known Activations