INDEX

Explanations

adverbs modifying an action

oai_token-act-pair · gpt-3.5-turbo

terms indicating exclusivity or singular focus

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

anth

-0.67

anon

-0.65

ylan

-0.63

yl

-0.61

itely

-0.59

enos

-0.59

LI

-0.58

asta

-0.58

 Leopard

-0.58

 vividly

-0.57

POSITIVE LOGITS

 reliant

1.01

 relying

0.97

 focused

0.90

foc

0.90

 responsible

0.85

 catering

0.83

 benefiting

0.81

 rely

0.79

 geared

0.79

 comprised

0.77

Activations Density 0.050%

adverbs modifying an action

terms indicating exclusivity or singular focus

No Comments

No Known Activations