INDEX

Explanations

words or parts of words ending in 'ie'

oai_token-act-pair · gpt-3.5-turbo

words that end with the suffix ‘ie’ or variations of that ending

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

INESS

-0.64

 Ended

-0.57

 Killing

-0.56

 "$:/

-0.55

 Oops

-0.55

 ACTIONS

-0.55

 boots

-0.54

 ecstasy

-0.54

 Wanted

-0.54

 killers

-0.53

POSITIVE LOGITS

lla

0.95

bert

0.94

ÃŁ

0.92

ppel

0.91

le

0.89

llo

0.88

ll

0.88

lect

0.87

bold

0.87

hart

0.87

Activations Density 0.099%

words or parts of words ending in 'ie'

words that end with the suffix ‘ie’ or variations of that ending

No Comments

No Known Activations