© 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. 2-RES-JB
    11. 3914
    Prev
    Next
    INDEX
    Explanations

    phrases related to large quantities of people, jobs, or money being impacted

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to large quantities or numbers, particularly in the context of economic or employment statistics

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ħĭ
    -0.74
     Cosponsors
    -0.67
     arrang
    -0.66
     misunder
    -0.66
    ĸļ
    -0.64
    iquette
    -0.64
     Ratio
    -0.62
     pairing
    -0.61
     Rabb
    -0.60
    SourceFile
    -0.59
    POSITIVE LOGITS
    perties
    0.79
     pounds
    0.76
     dollars
    0.71
     course
    0.67
    ohyd
    0.64
    icial
    0.62
     households
    0.62
     kilometers
    0.62
    pees
    0.62
     groceries
    0.61
    Activations Density 0.649%

    No Known Activations