INDEX

Explanations

words related to contradictions or opposition between different elements or ideas

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

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

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

*/(

-0.87

istically

-0.75

NetMessage

-0.75

sonian

-0.71

istic

-0.71

itatively

-0.69

oak

-0.69

LCS

-0.68

////////

-0.67

mares

-0.64

POSITIVE LOGITS

ption

0.90

ptions

0.84

 Aires

0.78

ghan

0.78

ventions

0.77

coni

0.76

ven

0.71

asca

0.71

xon

0.71

qual

0.69

Activations Density 8.910%

words related to contradictions or opposition between different elements or ideas

No Known Activations