© 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. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    5. GPT2-Small
    7. Residual Similar CE Local
    9. 6-RES_SCL-AJT
    11. 40326
    Prev
    Next
    INDEX
    Explanations

    references to computer codes and commands

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    neuronpedia/gpt2-small__res_scl-ajt/6-res_scl-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
    527
    +0.12
    0.5%
    149
    +0.11
    0.5%
    667
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    556
    +0.12
    0.13
    404
    +0.11
    0.12
    527
    +0.10
    0.03
    Negative Logits
     Corpus
    -0.68
    nesday
    -0.67
    arcity
    -0.65
    ournals
    -0.63
     Stras
    -0.63
     incap
    -0.61
     motivational
    -0.61
    bler
    -0.61
     fundament
    -0.60
     poles
    -0.60
    POSITIVE LOGITS
    achev
    0.93
    endi
    0.88
    heim
    0.81
    OOD
    0.81
    hea
    0.80
    tes
    0.80
    etric
    0.77
    gd
    0.76
    pg
    0.76
    INAL
    0.75
    Activations Density 19.341%

    No Known Activations