INDEX

Explanations

words related to physical locations or institutions

oai_token-act-pair · gpt-3.5-turbo

occurrences of the substring "ls" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 redistributed

-0.65

 partic

-0.63

 Boards

-0.62

 hemor

-0.61

TRE

-0.60

 Gerry

-0.59

 line

-0.59

Interstitial

-0.58

è¦ļéĨĴ

-0.58

 LINE

-0.57

POSITIVE LOGITS

creen

1.15

ls

1.03

ayers

0.96

hip

0.93

mas

0.93

pie

0.93

mith

0.92

aurus

0.91

peed

0.91

nyder

0.90

Activations Density 0.008%

words related to physical locations or institutions

occurrences of the substring "ls" in various contexts

No Comments

No Known Activations