© 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
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 4-RES-JB
    11. 10161
    Prev
    Next
    INDEX
    Explanations

    capital letters in Cyrillic script

    oai_token-act-pair · gpt-3.5-turbo

    characters or symbols specific to a particular encoding or language format

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

    No Comments

    Negative Logits
    Joy
    -0.76
    auga
    -0.72
    terson
    -0.67
    combe
    -0.66
     oun
    -0.62
    ierrez
    -0.62
     adolesc
    -0.62
     Riley
    -0.62
     Clover
    -0.60
     Spur
    -0.60
    POSITIVE LOGITS
    Ñĥ
    1.30
    а
    1.29
    оÐ
    1.28
    и
    1.26
    о
    1.25
    е
    1.21
    н
    1.16
    ÑĢ
    1.06
    к
    1.05
    Ñģ
    1.04
    Activations Density 0.008%

    No Known Activations