INDEX

Explanations

words related to advertisements and viewership

oai_token-act-pair · gpt-3.5-turbo

phrases that include expressions of gratitude or thanks

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

minist

-0.81

nown

-0.73

ESA

-0.69

 disagreement

-0.65

stress

-0.64

acter

-0.64

theless

-0.62

 denomin

-0.62

²¾

-0.62

epad

-0.61

POSITIVE LOGITS

umbnails

0.83

ijah

0.77

 Visit

0.75

 View

0.73

+)

0.70

 Reader

0.66

Interstitial

0.66

Kid

0.66

 viewer

0.65

 Featured

0.65

Activations Density 0.029%

words related to advertisements and viewership

phrases that include expressions of gratitude or thanks

No Comments

No Known Activations