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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlog/PodcastSlackPrivacy & 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. 0-RES-JB
    6. 3882
    Prev
    Next
    INDEX
    Explanations

    mentions of traffic congestion and related terms like "congested," "congestion relief," and "congestive heart failure."

    oai_token-act-pair · gpt-3.5-turbo

    terms related to congestion and traffic issues

    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
    ija
    -0.85
     Alchemist
    -0.78
    iron
    -0.76
    aith
    -0.72
    azi
    -0.72
    atech
    -0.72
     witch
    -0.69
     Witches
    -0.69
     Bir
    -0.69
     Bret
    -0.69
    POSITIVE LOGITS
     congestion
    2.80
     congest
    1.86
     densely
    1.35
     sparse
    1.31
     commute
    1.30
     planners
    1.26
     planner
    1.15
     Gillespie
    1.14
     contiguous
    1.13
     fragmentation
    1.12
    Activations Density 0.046%

    No Known Activations