INDEX

Explanations

phrases related to expressing thoughts or opinions

oai_token-act-pair · gpt-3.5-turbo

expressions of personal opinions and hypothetical reasoning

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

abases

-0.74

azar

-0.64

Lastly

-0.61

Ult

-0.61

Scroll

-0.61

emn

-0.60

gren

-0.59

nan

-0.59

oa

-0.59

avier

-0.59

POSITIVE LOGITS

 rather

0.75

 merely

0.68

uggest

0.65

¥µ

0.61

 indicative

0.61

INESS

0.61

POSE

0.60

 Rather

0.60

rather

0.58

rav

0.58

Activations Density 0.202%

phrases related to expressing thoughts or opinions

expressions of personal opinions and hypothetical reasoning

No Comments

No Known Activations