INDEX

Explanations

years mentioned in a text

oai_token-act-pair · gpt-3.5-turbo

references to specific years

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

acci

-0.81

 Skinner

-0.68

 maple

-0.66

vertisement

-0.65

bip

-0.63

meaning

-0.58

inance

-0.58

 diaper

-0.58

illet

-0.57

icone

-0.57

POSITIVE LOGITS

GV

0.86

Guard

0.70

âĢ¢âĢ¢

0.70

AMY

0.69

é¾į

0.68

TED

0.68

ãģ®æ

0.65

 Nicol

0.64

IVERS

0.64

Wars

0.64

Activations Density 0.176%

years mentioned in a text

references to specific years

No Comments

No Known Activations