© 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. 6529
    Prev
    Next
    INDEX
    Explanations

    prepositions and verbs indicating direction or viewpoint

    oai_token-act-pair · gpt-3.5-turbo

    the preposition "from" that indicates perspective or origin in a discussion

    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
    faced
    -0.74
    camp
    -0.70
    ifix
    -0.68
    irm
    -0.68
    busters
    -0.67
    reddits
    -0.67
    tailed
    -0.66
    isec
    -0.66
    rav
    -0.65
    Mask
    -0.64
    POSITIVE LOGITS
     afar
    1.84
     scratch
    1.32
     whence
    1.24
     standpoint
    1.04
     inception
    1.02
     outset
    1.02
     within
    1.00
     thence
    0.99
     inside
    0.97
    within
    0.90
    Activations Density 0.159%

    No Known Activations