INDEX

Explanations

words related to email subscription interactions

oai_token-act-pair · gpt-3.5-turbo

expressions of gratitude and subscription confirmations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

quartered

-0.75

mable

-0.67

etheless

-0.65

bilt

-0.62

livious

-0.61

urized

-0.61

etsk

-0.60

sonian

-0.60

 Ratt

-0.60

âĢİ

-0.58

POSITIVE LOGITS

 unlocks

0.75

interstitial

0.66

AY

0.66

google

0.63

Sorry

0.63

 subscribing

0.63

api

0.61

à¤

0.59

Suc

0.58

Buyable

0.58

Activations Density 0.030%

words related to email subscription interactions

expressions of gratitude and subscription confirmations

No Comments

No Known Activations