© 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. 8-RES-JB
    11. 20237
    Prev
    Next
    INDEX
    Explanations

    years and numerical values related to statistics or financial data

    oai_token-act-pair · gpt-3.5-turbo

    specific years and statistical data related to them

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     fusion
    -0.68
     masc
    -0.65
     bluff
    -0.64
     overlooking
    -0.61
     cha
    -0.61
     chess
    -0.60
     abras
    -0.60
     eccentric
    -0.60
     screws
    -0.59
     shortcuts
    -0.59
    POSITIVE LOGITS
     alone
    0.94
     compared
    0.90
    -'
    0.87
    adjusted
    0.86
     (%
    0.85
     GDP
    0.80
     onwards
    0.80
     Census
    0.80
     attributable
    0.77
     accounted
    0.77
    Activations Density 0.180%

    No Known Activations