INDEX

Explanations

phrases indicating a choice or alternative option

oai_token-act-pair · gpt-3.5-turbo

phrases or words indicating alternative perspectives or choices

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

urated

-0.77

oru

-0.70

usters

-0.70

osponsors

-0.67

anmar

-0.65

gur

-0.65

amia

-0.64

rys

-0.63

ardy

-0.63

rush

-0.62

POSITIVE LOGITS

 sucks

0.75

 imaginable

0.72

 else

0.70

 depends

0.68

 thereafter

0.67

 construed

0.65

soever

0.65

 preferable

0.65

>>>>

0.64

event

0.63

Activations Density 0.039%

phrases indicating a choice or alternative option

phrases or words indicating alternative perspectives or choices

No Comments

No Known Activations