INDEX

Explanations

proper nouns or names ending in 'ien'

oai_token-act-pair · gpt-3.5-turbo

terms associated with scientific classifications or identifiers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

è¦ļéĨĴ

-0.92

ramid

-0.89

atform

-0.87

rooms

-0.72

agna

-0.68

taboola

-0.66

tm

-0.64

hematic

-0.64

 mids

-0.64

 oats

-0.63

POSITIVE LOGITS

vironment

1.18

cia

0.96

ews

0.89

hao

0.87

emies

0.82

cing

0.82

emy

0.81

flix

0.80

cest

0.79

icidal

0.78

Activations Density 0.022%

proper nouns or names ending in 'ien'

terms associated with scientific classifications or identifiers

No Comments

No Known Activations