© 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. 480
    Prev
    Next
    INDEX
    Explanations

    phrases starting with 'again,' followed by various contexts and statements, such as descriptions of actions and opinions

    oai_token-act-pair · gpt-3.5-turbo

    the repetition of phrases or concepts throughout the text

    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
    emo
    -0.69
     nonetheless
    -0.62
     nevertheless
    -0.61
    vice
    -0.60
     umbrella
    -0.58
     Built
    -0.58
     Reaction
    -0.57
     precursor
    -0.57
     Cooldown
    -0.56
     Ki
    -0.56
    POSITIVE LOGITS
    ukong
    0.75
    nces
    0.71
    aughed
    0.71
    atever
    0.70
    dor
    0.69
     Nieto
    0.67
    ilaterally
    0.64
    orks
    0.63
    ãĥ£
    0.63
    ãĤŃ
    0.63
    Activations Density 0.132%

    No Known Activations