Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    3. GPT2-Small
    4. Residual Similar CE E2E Recon
    5. 6-RES_SCEFR-AJT
    6. 443
    Prev
    Next
    INDEX
    Explanations

    economic and financial information

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    neuronpedia/gpt2-small__res_scefr-ajt/6-res_scefr-ajt
    Prompts (Dashboard)
    12,288 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    46,080
    Data Type
    torch.float32
    Hook Point
    blocks.6.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    apollo-research/Skylion007-openwebtext-tokenizer-gpt2
    Hook Point Layer
    6
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    578
    +0.20
    0.6%
    314
    +0.17
    0.5%
    195
    +0.17
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    205
    +0.20
    0.07
    578
    +0.17
    0.05
    303
    +0.17
    0.06
    Negative Logits
     exerc
    -1.37
     Gors
    -1.29
     Flavoring
    -1.28
     constitu
    -1.25
    umbn
    -1.22
     Pend
    -1.16
     Vid
    -1.15
     Moody
    -1.13
     Bere
    -1.12
     partName
    -1.12
    POSITIVE LOGITS
    $$
    2.00
    $$$$
    1.90
    -$
    1.47
    USD
    1.46
    ($
    1.46
    GOP
    1.37
     Donation
    1.36
    dollar
    1.36
    false
    1.35
    EDIT
    1.34
    Activations Density 0.498%

    No Known Activations