INDEX

Explanations

names or labels

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate listing or naming items or examples

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ysc

-0.69

entimes

-0.68

loo

-0.67

issance

-0.67

 childbirth

-0.63

 deterior

-0.63

tail

-0.60

 propelled

-0.60

Returns

-0.60

depth

-0.60

POSITIVE LOGITS

 culprit

0.86

 names

0.85

 specific

0.81

 perpetrators

0.76

NCT

0.74

 blame

0.74

 Names

0.73

GROUP

0.72

 perpetrator

0.71

 particular

0.71

Activations Density 0.317%

names or labels

phrases that indicate listing or naming items or examples

No Comments

No Known Activations