INDEX

Explanations

assertions or predictions about future events

oai_token-act-pair · gpt-3.5-turbo

phrases indicating existence or occurrence of something

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 Industries

-0.64

zer

-0.64

rix

-0.62

strip

-0.59

era

-0.58

azar

-0.57

plug

-0.56

berman

-0.56

ital

-0.56

arta

-0.56

POSITIVE LOGITS

 plenty

1.17

 repercussions

1.04

 exceptions

0.99

no

0.97

 fewer

0.95

 ample

0.91

 enough

0.91

 lots

0.90

 consequences

0.90

 occasions

0.87

Activations Density 0.058%

assertions or predictions about future events

phrases indicating existence or occurrence of something

No Comments

No Known Activations