INDEX

Explanations

phrases indicating contrast or difference

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate differences or variations between subjects

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

icide

-0.71

VICE

-0.67

indust

-0.66

record

-0.64

phies

-0.64

vice

-0.63

wind

-0.63

ongyang

-0.62

stop

-0.62

ocious

-0.62

POSITIVE LOGITS

Different

0.78

 differing

0.77

":"/

0.77

 Differences

0.77

 personalities

0.71

 depending

0.70

 timelines

0.70

 Original

0.68

 Same

0.67

Race

0.67

Activations Density 0.583%

phrases indicating contrast or difference

phrases that indicate differences or variations between subjects

No Comments

No Known Activations