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. 5-RES-JB
    6. 19635
    Prev
    Next
    INDEX
    Explanations

    references to applications (apps)

    oai_token-act-pair · gpt-3.5-turbo

    references to applications in the context of technology

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 5-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    5
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    rique
    -0.86
    neck
    -0.70
    liction
    -0.70
    bilt
    -0.68
    few
    -0.66
     Patriot
    -0.66
    construct
    -0.66
    orr
    -0.61
    isen
    -0.60
     induction
    -0.60
    POSITIVE LOGITS
    ynthesis
    0.95
    uggest
    0.84
     apps
    0.82
    ystem
    0.81
    linger
    0.79
     downloaded
    0.79
    paces
    0.79
    cape
    0.77
     installed
    0.76
    pring
    0.75
    Activations Density 0.046%

    No Known Activations