INDEX

Explanations

names of characters or locations ending in 'ang' or 'ong'

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

rees

-0.69

ively

-0.68

ngth

-0.67

ional

-0.65

ortun

-0.62

ibles

-0.60

alion

-0.59

uously

-0.59

iaries

-0.58

ives

-0.58

POSITIVE LOGITS

hound

0.61

lda

0.60

mberg

0.59

 Olymp

0.58

wolf

0.58

ford

0.58

gang

0.56

ORPG

0.54

 Yang

0.53

PAC

0.53

Activations Density 7.703%

names of characters or locations ending in 'ang' or 'ong'

No Comments

No Known Activations