INDEX

Explanations

phrases expressing doubt or skepticism

oai_token-act-pair · gpt-3.5-turbo

negations and the dismissal of previously mentioned ideas

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

illac

-0.67

oir

-0.66

uthor

-0.64

APH

-0.64

Â¯Â¯Â¯Â¯Â¯Â¯Â¯Â¯

-0.64

assic

-0.63

oise

-0.61

airs

-0.60

utic

-0.59

hyde

-0.59

POSITIVE LOGITS

etheless

1.11

 despair

0.91

 unreasonable

0.86

 neglect

0.86

 detract

0.85

 overlooked

0.80

 discouraged

0.80

theless

0.78

 denying

0.78

 nonetheless

0.78

Activations Density 0.272%

phrases expressing doubt or skepticism

negations and the dismissal of previously mentioned ideas

No Comments

No Known Activations