INDEX

Explanations

phrases related to length or duration

oai_token-act-pair · gpt-3.5-turbo

the word "longer" in various contexts

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

Stars

-0.84

elf

-0.81

IRO

-0.77

Ro

-0.73

Sov

-0.72

ery

-0.72

Always

-0.71

mir

-0.69

 Coordinator

-0.69

eers

-0.68

POSITIVE LOGITS

 than

1.27

than

1.04

 Than

1.00

 paced

0.89

 periods

0.87

 longer

0.85

 distances

0.85

 stretches

0.84

 lasting

0.84

neck

0.77

Activations Density 0.022%

phrases related to length or duration

the word "longer" in various contexts

No Comments

No Known Activations