INDEX

Explanations

scientific terms and technical descriptions

oai_token-act-pair · gpt-3.5-turbo

nouns related to groups and collections of individuals or entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

ebus

-0.68

ĨĴ

-0.67

 Alone

-0.66

 Trafford

-0.65

 Shed

-0.65

 Tire

-0.64

adish

-0.63

%]

-0.62

 Rampage

-0.61

 Anniversary

-0.60

POSITIVE LOGITS

 converge

0.87

 including

0.82

 ranging

0.73

izable

0.72

votes

0.71

 vying

0.71

 relating

0.71

 conver

0.70

isable

0.70

 involving

0.68

Activations Density 0.454%

scientific terms and technical descriptions

nouns related to groups and collections of individuals or entities

No Comments

No Known Activations