INDEX

Explanations

phrases related to achievements and successes

oai_token-act-pair · gpt-3.5-turbo

negative assessments of products and experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

Coun

-0.70

 ÂŃ

-0.65

Connell

-0.65

uble

-0.63

Keefe

-0.63

Gov

-0.61

agogue

-0.60

Newsletter

-0.60

itored

-0.60

heit

-0.59

POSITIVE LOGITS

 tremend

1.10

 alot

1.01

 phenomenal

0.99

 amazing

0.96

 HUGE

0.95

 greatness

0.93

 tremendous

0.92

 great

0.89

 fantastic

0.89

 incredible

0.88

Activations Density 0.567%

phrases related to achievements and successes

negative assessments of products and experiences

No Comments

No Known Activations