© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    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. 14755
    Prev
    Next
    INDEX
    Explanations

    words in a non-Latin script, possibly Cyrillic

    oai_token-act-pair · gpt-3.5-turbo

    characters or letters from a specific non-Latin script

    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
    ometimes
    -0.82
     GOODMAN
    -0.77
    oppable
    -0.76
    DragonMagazine
    -0.74
    bucks
    -0.71
    ierrez
    -0.71
    emort
    -0.70
    emonium
    -0.69
    otaur
    -0.69
    agonist
    -0.68
    POSITIVE LOGITS
    Ñı
    1.19
    н
    1.18
    ÑĤ
    1.08
    ¶
    1.08
    м
    1.07
    в
    1.07
    Ð
    1.06
    ¾
    1.06
    л
    1.06
    Ñ
    1.04
    Activations Density 0.005%

    No Known Activations