INDEX

Explanations

the word "is" followed by an adjective or noun phrase, often in the context of questioning or criticism

oai_token-act-pair · gpt-3.5-turbo

questions that seek clarification or explanation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ptoms

-0.75

brates

-0.75

eers

-0.74

aneers

-0.73

weights

-0.70

tails

-0.69

 credits

-0.68

 Sabres

-0.68

 IMAGES

-0.67

details

-0.67

POSITIVE LOGITS

olated

1.37

olation

1.34

olate

1.16

htar

1.04

abella

0.96

omorphic

0.91

earch

0.88

peria

0.85

terness

0.84

ometric

0.83

Activations Density 0.130%

the word "is" followed by an adjective or noun phrase, often in the context of questioning or criticism

questions that seek clarification or explanation

No Comments

No Known Activations