© 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. 4-RES-JB
    11. 593
    Prev
    Next
    INDEX
    Explanations

    words ending in 'h' with a high activation value

    oai_token-act-pair · gpt-3.5-turbo

    instances of the symbol 'h'

    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
    ãĤ´ãĥ³
    -0.91
    ãĥ¯
    -0.89
    éĹĺ
    -0.87
    ãĥ¼ãĥĨãĤ£
    -0.74
    ortium
    -0.74
    ãĥ´ãĤ¡
    -0.73
    DragonMagazine
    -0.73
    EStream
    -0.69
    ierrez
    -0.67
    ãĥ¼ãĤ¯
    -0.67
    POSITIVE LOGITS
    oused
    1.27
    ousing
    1.17
    awk
    1.14
    ulk
    1.12
    acking
    1.12
    anging
    1.09
    idd
    1.08
    annah
    1.06
    olly
    1.05
    oney
    1.05
    Activations Density 0.018%

    No Known Activations