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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & 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. 8-RES-JB
    6. 2884
    Prev
    Next
    INDEX
    Explanations

    words related to a specific geographical location, "Dav"

    oai_token-act-pair · gpt-3.5-turbo

    references to the location "Davao" or names containing "Dav."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     carriers
    -0.69
     occup
    -0.67
    ACTED
    -0.65
     alpha
    -0.64
     crochet
    -0.62
     explosives
    -0.61
     ultr
    -0.60
    TRY
    -0.60
     caption
    -0.60
     pref
    -0.60
    POSITIVE LOGITS
    isson
    1.03
    ict
    1.01
    ildo
    1.00
    enture
    0.99
    inas
    0.99
    rolet
    0.98
    inelli
    0.97
    iral
    0.96
    illon
    0.95
    olla
    0.95
    Activations Density 0.027%

    No Known Activations