INDEX

Explanations

numbers followed by the word 'fold'

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

Not in Any Lists

No Comments

Negative Logits

APS

-0.64

igslist

-0.61

anu

-0.59

ampires

-0.59

enda

-0.59

omo

-0.58

onde

-0.57

 Surv

-0.57

acea

-0.57

EA

-0.57

POSITIVE LOGITS

fold

0.87

 increase

0.74

 proportions

0.62

shr

0.61

theless

0.61

inqu

0.61

 fold

0.60

 reduction

0.59

charge

0.58

edly

0.57

Activations Density 7.869%

numbers followed by the word 'fold'

No Comments

No Known Activations