INDEX

Explanations

adjectives describing different kinds of purposes or motives

oai_token-act-pair · gpt-3.5-turbo

terms related to politics and motivations for actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

etheless

-0.92

gall

-0.73

anon

-0.72

bek

-0.66

glas

-0.63

Gar

-0.60

cheat

-0.60

Pad

-0.59

 Accessed

-0.58

rogens

-0.58

POSITIVE LOGITS

 purposes

2.16

 sake

2.02

 reasons

1.51

 purpose

1.32

 reason

1.12

ummies

1.06

 Reasons

0.93

aughs

0.89

 convenience

0.87

 gratification

0.85

Activations Density 0.843%

adjectives describing different kinds of purposes or motives

terms related to politics and motivations for actions

No Comments

No Known Activations