INDEX

Explanations

the word "or" highlighted in different contexts

oai_token-act-pair · gpt-3.5-turbo

instances of the word "or" indicating choice or alternatives

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

uay

-0.69

 Scare

-0.65

OTAL

-0.64

 Outbreak

-0.61

onday

-0.61

ivid

-0.60

 Showdown

-0.59

INT

-0.59

 Pigs

-0.58

eared

-0.58

POSITIVE LOGITS

chid

1.05

lando

1.00

Else

0.93

chard

0.92

thodox

0.92

acle

0.88

phan

0.86

adian

0.85

anges

0.84

 Marketable

0.82

Activations Density 0.016%

the word "or" highlighted in different contexts

instances of the word "or" indicating choice or alternatives

No Comments

No Known Activations