© 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. 3115
    Prev
    Next
    INDEX
    Explanations

    words or phrases related to technical instructions or coding commands

    oai_token-act-pair · gpt-3.5-turbo

    technical jargon and terms related to technology and data management

    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
     XIII
    -0.67
     Trident
    -0.63
     life
    -0.62
     Neptune
    -0.62
     Lib
    -0.61
     Hat
    -0.60
     Keystone
    -0.60
     Tart
    -0.60
     Breed
    -0.60
     Fu
    -0.60
    POSITIVE LOGITS
    posted
    0.84
    è¦ļéĨĴ
    0.84
    answered
    0.83
    cause
    0.82
    accompan
    0.79
    TPPStreamerBot
    0.79
    é»Ĵ
    0.78
    inducing
    0.76
    except
    0.75
    ãĥ´
    0.75
    Activations Density 0.256%

    No Known Activations