INDEX

Explanations

terms indicating additional information or emphasis

oai_token-act-pair · gpt-3.5-turbo

transitional phrases indicating elaboration or addition in arguments

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

"},"

-0.73

crow

-0.60

ICAN

-0.55

/"

-0.55

ggles

-0.54

prus

-0.53

rals

-0.50

packs

-0.48

DAY

-0.48

 VIDEOS

-0.48

POSITIVE LOGITS

0.84

,.

0.71

 guiActiveUn

0.69

,,

0.68

.,

0.67

,...

0.65

!,

0.63

entimes

0.63

é¾įåĸļå£«

0.59

 notwithstanding

0.58

Activations Density 0.064%

terms indicating additional information or emphasis

transitional phrases indicating elaboration or addition in arguments

No Comments

No Known Activations