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. 18644
    Prev
    Next
    INDEX
    Explanations

    words related to things or activities being done with effort or force, such as "trying", "training", "tracking"

    oai_token-act-pair · gpt-3.5-turbo

    instances of the substring "tr."

    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
    ELS
    -0.85
    LEY
    -0.74
     Mechdragon
    -0.72
    ãģ®éŃĶ
    -0.70
    æĸ¹
    -0.69
    WARE
    -0.68
    eers
    -0.67
    ELD
    -0.66
    MODE
    -0.66
    ARDS
    -0.65
    POSITIVE LOGITS
    umps
    1.30
    ainer
    1.17
    acer
    1.11
    unks
    1.08
    usted
    1.08
    iceps
    1.05
    ained
    1.04
    ains
    1.02
    icol
    1.02
    amm
    1.02
    Activations Density 0.009%

    No Known Activations