INDEX

Explanations

the term "RSS"

oai_token-act-pair · gpt-3.5-turbo

references to RSS feeds

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

È

-0.79

 Hurricanes

-0.77

perty

-0.75

esville

-0.75

ecause

-0.73

eu

-0.70

cellence

-0.70

tesy

-0.68

 Kardashian

-0.65

lain

-0.61

POSITIVE LOGITS

 feeds

1.05

rss

0.97

RSS

0.95

 feed

0.89

adish

0.85

 Feed

0.81

 mosqu

0.80

DP

0.79

ICAL

0.73

 cows

0.70

Activations Density 0.016%

the term "RSS"

references to RSS feeds

No Comments

No Known Activations