© 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. 9866
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "Bal" with various numerical suffixes

    oai_token-act-pair · gpt-3.5-turbo

    repeated mentions of the term "Bal" in various contexts

    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
    ä
    -0.73
    æĸ¹
    -0.70
    ä½ľ
    -0.68
     Patriot
    -0.68
     plural
    -0.66
    DAQ
    -0.65
     Chaser
    -0.65
    HER
    -0.64
    CRIP
    -0.62
     Doodle
    -0.60
    POSITIVE LOGITS
    boa
    1.27
    zac
    1.09
    ancing
    1.08
    anca
    1.04
    ivari
    0.99
    aji
    0.98
    uca
    0.96
    inese
    0.95
    asar
    0.95
    anche
    0.93
    Activations Density 0.022%

    No Known Activations