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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 1-RES-JB
    11. 3763
    Prev
    Next
    INDEX
    Explanations

    numbers or numerical patterns

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the sequence "enty" and variations of "ause" and "vation" which suggest a focus on causative or descriptive phrases

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    è¦ļéĨĴ
    -0.72
    Amazing
    -0.64
    oub
    -0.63
     informants
    -0.61
    âĢ¢âĢ¢âĢ¢âĢ¢
    -0.61
     breathtaking
    -0.60
    Legend
    -0.58
     Vortex
    -0.58
    TEXTURE
    -0.58
    ĸļ士
    -0.58
    POSITIVE LOGITS
    enty
    0.96
    gypt
    0.86
    ause
    0.85
    vation
    0.84
    culosis
    0.81
    idth
    0.77
    entimes
    0.77
    ocene
    0.74
    raviolet
    0.74
    neum
    0.73
    Activations Density 0.020%

    No Known Activations