INDEX

Explanations

phrases containing the phrase "no-"

oai_token-act-pair · gpt-3.5-turbo

negative constructs or phrases

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

 Reloaded

-0.77

gypt

-0.70

 Reborn

-0.68

APTER

-0.66

çīĪ

-0.66

 Rouge

-0.66

 vind

-0.65

 Fork

-0.65

 Royale

-0.64

 Sparrow

-0.63

POSITIVE LOGITS

brainer

1.21

matter

1.08

interest

1.02

notice

1.01

smoking

1.01

exc

0.98

repeat

0.98

issue

0.97

contact

0.97

strings

0.97

Activations Density 0.026%

phrases containing the phrase "no-"

negative constructs or phrases

No Comments

No Known Activations