INDEX

Explanations

text related to incorporating instructions or elements into something, possibly in a technical or creative context

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

uku

-0.63

ritz

-0.62

entimes

-0.61

dan

-0.61

zai

-0.61

yip

-0.60

achi

-0.58

die

-0.58

ebted

-0.58

iolet

-0.58

POSITIVE LOGITS

 adulthood

0.77

 society

0.60

 existence

0.59

 mainland

0.58

 orbit

0.58

 Valhalla

0.57

 academia

0.57

 Unity

0.55

 guiActiveUnfocused

0.55

ob

0.54

Activations Density 12.056%

text related to incorporating instructions or elements into something, possibly in a technical or creative context

No Comments

No Known Activations