© 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. 8473
    Prev
    Next
    INDEX
    Explanations

    words related to the sound "hum."

    oai_token-act-pair · gpt-3.5-turbo

    forms of the word "hum" and its variations, indicating a focus on themes of humility or humor

    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
    folios
    -0.73
    okemon
    -0.67
    eanor
    -0.66
    hyde
    -0.65
    Constructed
    -0.63
     Princ
    -0.60
    ع
    -0.60
    ãĥī
    -0.59
    holders
    -0.58
    holder
    -0.58
    POSITIVE LOGITS
    bly
    1.19
    bling
    1.11
    bled
    1.09
     hum
    1.07
    eless
    1.04
    bler
    1.00
    bug
    0.99
    ingly
    0.94
    bles
    0.93
    ulus
    0.91
    Activations Density 0.012%

    No Known Activations