INDEX

Explanations

prepositional phrases indicating a specific type of action or behavior

oai_token-act-pair · gpt-3.5-turbo

phrases that express various types of attention or critique

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

 Pigs

-0.75

 Crush

-0.72

gow

-0.70

pots

-0.66

 Sands

-0.65

 Rocks

-0.64

gor

-0.64

APS

-0.64

mates

-0.62

ours

-0.61

POSITIVE LOGITS

 thing

1.02

 scenario

0.81

 behavior

0.75

 stuff

0.73

 nonsense

0.72

 activity

0.71

 attrition

0.70

 kindred

0.69

tnc

0.69

 fate

0.68

Activations Density 0.033%

prepositional phrases indicating a specific type of action or behavior

phrases that express various types of attention or critique

No Comments

No Known Activations