© 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. 3141
    Prev
    Next
    INDEX
    Explanations

    instances of the word 'separate'

    oai_token-act-pair · gpt-3.5-turbo

    references to things being categorized as separate entities or groups

    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
    Ö¼
    -0.71
    eno
    -0.68
     onwards
    -0.65
    enos
    -0.64
    amac
    -0.63
    ENE
    -0.63
     Beans
    -0.63
    older
    -0.62
    Des
    -0.61
    eny
    -0.61
    POSITIVE LOGITS
     separate
    0.81
     entity
    0.81
    iates
    0.80
    ively
    0.77
    worldly
    0.77
    iating
    0.77
    ances
    0.75
    bie
    0.75
     piece
    0.73
     stint
    0.72
    Activations Density 0.037%

    No Known Activations