INDEX

Explanations

phrases related to uncertainty and speculation

oai_token-act-pair · gpt-3.5-turbo

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

ModLoader

-0.74

 Ratio

-0.74

ĨĴ

-0.72

Template

-0.71

©¶æ¥µ

-0.71

è¦ļéĨĴ

-0.70

ELD

-0.70

 Calculator

-0.69

MAP

-0.69

¿½

-0.69

POSITIVE LOGITS

 react

1.15

 bother

1.09

 approve

1.06

 appreciate

1.02

 come

1.01

 realize

1.00

 disapprove

1.00

 wanna

0.99

 notice

0.99

 cooperate

0.98

Activations Density 0.421%

phrases related to uncertainty and speculation

No Comments

No Known Activations