INDEX

Explanations

phrases related to experiences or occurrences that are significant or meaningful to the speaker

oai_token-act-pair · gpt-3.5-turbo

expressions of positive emotions and experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

pmwiki

-0.73

 beware

-0.72

 accuses

-0.71

 complains

-0.69

 disapprove

-0.69

 accuse

-0.68

 objections

-0.68

 complaints

-0.67

 disappro

-0.66

inyl

-0.65

POSITIVE LOGITS

 FANT

1.20

 fantastic

1.15

 awesome

1.14

 wonderful

1.11

 amazing

1.10

 GREAT

1.07

 invaluable

1.06

 priceless

1.04

 incredible

1.01

 terrific

0.99

Activations Density 0.338%

phrases related to experiences or occurrences that are significant or meaningful to the speaker

expressions of positive emotions and experiences

No Comments

No Known Activations