© 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. 10262
    Prev
    Next
    INDEX
    Explanations

    phrases related to applying pressure or fitting something tightly into a limited space

    oai_token-act-pair · gpt-3.5-turbo

    the verb "squeeze" and its variations

    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
    ellery
    -0.82
    amation
    -0.79
    brance
    -0.75
    quartered
    -0.73
    igel
    -0.73
    etermination
    -0.73
    etheless
    -0.72
    rust
    -0.72
    unc
    -0.71
    ibel
    -0.71
    POSITIVE LOGITS
     squeeze
    1.11
     squee
    1.04
     squeezing
    1.02
     squeezed
    0.99
    hift
    0.79
    cules
    0.76
    weed
    0.72
     Doodle
    0.71
     pedal
    0.70
     necks
    0.70
    Activations Density 0.012%

    No Known Activations