INDEX

Explanations

detailed observations or insights in textual information

oai_token-act-pair · gpt-3.5-turbo

phrases indicating the observation or discovery of something

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

awar

-0.75

)]

-0.74

ovie

-0.73

oustic

-0.69

ffen

-0.66

youtube

-0.65

ajor

-0.65

iership

-0.65

orean

-0.64

nai

-0.63

POSITIVE LOGITS

 plenty

1.05

 numerous

1.01

 myriad

0.97

 that

0.95

 countless

0.93

 something

0.91

 dozens

0.91

 nothing

0.90

 lots

0.90

 innumerable

0.89

Activations Density 0.173%

detailed observations or insights in textual information

phrases indicating the observation or discovery of something

No Comments

No Known Activations