© 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. 3612
    Prev
    Next
    INDEX
    Explanations

    the word "or" being used as a conjunction

    oai_token-act-pair · gpt-3.5-turbo

    conjunctions and the concept of inclusion or alternation

    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
    anchester
    -0.78
    emet
    -0.67
    wolves
    -0.66
    grave
    -0.64
    conserv
    -0.59
    retty
    -0.58
    inea
    -0.58
    GMT
    -0.56
    furt
    -0.56
    wordpress
    -0.55
    POSITIVE LOGITS
    osterone
    0.78
    alore
    0.68
    IRED
    0.64
    ption
    0.63
     Shotgun
    0.61
    %"
    0.59
    tsky
    0.59
    iverse
    0.58
    issan
    0.58
    ĪĴ
    0.58
    Activations Density 0.257%

    No Known Activations