INDEX

Explanations

mentions of something new or recently discovered

oai_token-act-pair · gpt-3.5-turbo

the word "newly" and its variations

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

ĺħ

-0.75

orem

-0.75

ocene

-0.74

raints

-0.73

rice

-0.72

 sqor

-0.72

retty

-0.69

 Citation

-0.68

ashtra

-0.68

igraph

-0.68

POSITIVE LOGITS

wed

1.01

bie

0.90

 appointed

0.89

ãĤ»

0.82

 liberated

0.82

foundland

0.80

cedented

0.79

 acquired

0.79

 mint

0.78

 dubbed

0.78

Activations Density 0.016%

mentions of something new or recently discovered

the word "newly" and its variations

No Comments

No Known Activations