INDEX

Explanations

words related to ignoring or being ignored

oai_token-act-pair · gpt-3.5-turbo

instances of the word "ignore" and its variations in relation to various subjects or issues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

amide

-0.66

ickr

-0.66

uliffe

-0.65

zar

-0.65

rim

-0.65

uster

-0.64

emetery

-0.62

creation

-0.62

unal

-0.61

ccording

-0.60

POSITIVE LOGITS

ibly

0.96

ibility

0.71

illy

0.70

 aside

0.68

lessly

0.67

fulness

0.62

 Sakuya

0.61

 SERV

0.61

uously

0.60

ibilities

0.60

Activations Density 0.033%

words related to ignoring or being ignored

instances of the word "ignore" and its variations in relation to various subjects or issues

No Comments

No Known Activations