© 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. 6031
    Prev
    Next
    INDEX
    Explanations

    words related to breaking or disrupting something

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "break" in various contexts

    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
    imental
    -0.69
    ulp
    -0.66
    ifice
    -0.64
    metics
    -0.64
    omorphic
    -0.63
    00200000
    -0.61
    ogram
    -0.61
    rica
    -0.59
    gel
    -0.59
    velength
    -0.57
    POSITIVE LOGITS
    neck
    1.09
    fast
    0.99
    down
    0.96
    away
    0.94
    breakers
    0.91
    downs
    0.91
    break
    0.90
    points
    0.86
    breaks
    0.85
    breaking
    0.80
    Activations Density 0.026%

    No Known Activations