INDEX

Explanations

questions beginning with "Why."

oai_token-act-pair · gpt-3.5-turbo

rhetorical questions that challenge reasoning or assumptions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

interstitial

-0.66

è¦ļéĨĴ

-0.65

ILY

-0.63

eki

-0.60

unal

-0.60

aukee

-0.60

iece

-0.60

ipes

-0.59

apsed

-0.59

tips

-0.59

POSITIVE LOGITS

 bother

1.19

 shouldn

1.07

 wouldn

1.05

 aren

1.04

 didn

1.01

did

1.00

 does

0.99

 hasn

0.98

don

0.95

 should

0.93

Activations Density 0.044%

questions beginning with "Why."

rhetorical questions that challenge reasoning or assumptions

No Comments

No Known Activations