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. 3-RES-JB
    6. 13974
    Prev
    Next
    INDEX
    Explanations

    keywords related to beacons, specific locations ("Vale," "Ashe," "Bunker"), and a character name ("Cloak")

    oai_token-act-pair · gpt-3.5-turbo

    references to specific organizations, places, or characters

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ivan
    -0.88
    afort
    -0.77
    tery
    -0.76
    ktop
    -0.73
    surv
    -0.70
    Lenin
    -0.70
    andro
    -0.69
    tti
    -0.68
    ppe
    -0.67
    ulu
    -0.67
    POSITIVE LOGITS
     Beacon
    1.13
     bom
    0.80
     beacon
    0.74
    zeb
    0.71
    ãĤ¤ãĥĪ
    0.70
     Ridge
    0.69
     exting
    0.69
     Falls
    0.68
     Triangle
    0.68
     Academy
    0.66
    Activations Density 0.012%

    No Known Activations