© 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. 3-RES-JB
    11. 13610
    Prev
    Next
    INDEX
    Explanations

    mentions of specific fictional or real-world entities

    oai_token-act-pair · gpt-3.5-turbo

    the letter 'Ļ' represented in the text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     mathemat
    -0.78
     fortun
    -0.74
     imitation
    -0.67
     carbohyd
    -0.64
     Seym
    -0.64
    ramid
    -0.63
     pse
    -0.62
     popup
    -0.60
    awaru
    -0.60
     elig
    -0.60
    POSITIVE LOGITS
    ï¸ı
    1.06
    VICE
    0.78
    VERTISEMENT
    0.78
    İ
    0.77
    hall
    0.77
    女
    0.77
    âĢ¢âĢ¢âĢ¢âĢ¢
    0.76
    ½
    0.76
    ï¸
    0.75
    ULE
    0.74
    Activations Density 0.437%

    No Known Activations