© 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. 4-RES-JB
    11. 6172
    Prev
    Next
    INDEX
    Explanations

    references to Samsung Galaxy Note models

    oai_token-act-pair · gpt-3.5-turbo

    references to specific model names of smartphones, particularly Samsung Galaxy Note variants

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

    No Comments

    Negative Logits
     begg
    -0.79
    lehem
    -0.73
    é¾
    -0.72
    Interstitial
    -0.71
    ritch
    -0.70
    heast
    -0.67
    employment
    -0.67
    estern
    -0.64
    reek
    -0.63
    inates
    -0.63
    POSITIVE LOGITS
    book
    0.98
    BOOK
    0.89
    books
    0.89
    Pad
    0.88
    Maker
    0.82
    manship
    0.82
     ®
    0.76
    maker
    0.74
     Plus
    0.73
     handset
    0.73
    Activations Density 0.015%

    No Known Activations