© 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. 14701
    Prev
    Next
    INDEX
    Explanations

    mentions of specific periods of time, particularly weeks

    oai_token-act-pair · gpt-3.5-turbo

    references to the concept of a "week" 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
    emort
    -0.69
    ected
    -0.65
    awei
    -0.65
     Tigers
    -0.64
    ashtra
    -0.64
    itars
    -0.63
    inav
    -0.61
    ifice
    -0.59
    Frameworks
    -0.59
    cellence
    -0.59
    POSITIVE LOGITS
    days
    1.38
    night
    1.20
    DAY
    1.01
    iversary
    0.92
    lies
    0.85
    day
    0.82
    trip
    0.77
    aday
    0.75
    hof
    0.75
    orrow
    0.75
    Activations Density 0.043%

    No Known Activations