INDEX

Explanations

phrases related to online content consumption

oai_token-act-pair · gpt-3.5-turbo

phrases expressing gratitude or appreciation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

issan

-0.76

FTWARE

-0.70

 âĵĺ

-0.66

pora

-0.65

oft

-0.63

 embodiment

-0.62

WARE

-0.62

âĵĺ

-0.61

 corps

-0.61

ledged

-0.61

POSITIVE LOGITS

à¹

0.75

.)

0.72

phy

0.70

 Vaugh

0.65

ocks

0.63

ABC

0.63

aily

0.61

âĢ¦)

0.61

 intervened

0.61

yon

0.60

Activations Density 0.064%

phrases related to online content consumption

phrases expressing gratitude or appreciation

No Comments

No Known Activations