INDEX

Explanations

phrases related to removal or elimination

oai_token-act-pair · gpt-3.5-turbo

instances of the word "remove" and its related context

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

hof

-0.67

SPONSORED

-0.67

 Soup

-0.66

Launch

-0.65

 Bout

-0.65

lished

-0.63

é¾įåĸļå£«

-0.63

INFO

-0.61

swer

-0.60

ificent

-0.60

POSITIVE LOGITS

 altogether

1.07

 barrier

0.90

 barriers

0.85

 restraints

0.84

 veil

0.84

 entirely

0.83

 clutter

0.81

 pesky

0.79

 unwanted

0.78

 liability

0.76

Activations Density 0.340%

phrases related to removal or elimination

instances of the word "remove" and its related context

No Comments

No Known Activations