INDEX

Explanations

keywords related to concepts, plans, or proposals

oai_token-act-pair · gpt-3.5-turbo

phrases centered around the concept of ideas and their implications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

Ago

-0.70

hma

-0.70

Oper

-0.69

çĦ

-0.66

 Peaks

-0.63

wald

-0.62

 trophies

-0.62

eworthy

-0.60

 Coverage

-0.60

pex

-0.60

POSITIVE LOGITS

ually

1.05

 idea

0.80

 moot

0.80

atical

0.77

reprene

0.76

 hatched

0.75

orial

0.73

ual

0.70

matic

0.69

 reim

0.68

Activations Density 0.049%

keywords related to concepts, plans, or proposals

phrases centered around the concept of ideas and their implications

No Comments

No Known Activations