INDEX

Explanations

terms or phrases used to describe or categorize concepts

oai_token-act-pair · gpt-3.5-turbo

specific terminology and technical jargon

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

tsky

-0.69

thumbnails

-0.65

oppable

-0.63

phis

-0.61

aukee

-0.61

uala

-0.61

 recip

-0.60

 unsus

-0.59

igators

-0.58

illance

-0.58

POSITIVE LOGITS

 coined

1.24

 referring

1.08

 synonymous

1.04

ciation

1.00

 meaning

0.99

 shorthand

0.94

 describing

0.93

 loosely

0.91

 refers

0.86

 used

0.86

Activations Density 0.104%

terms or phrases used to describe or categorize concepts

specific terminology and technical jargon

No Comments

No Known Activations