INDEX

Explanations

specific keywords related to the origins or initial history of something

oai_token-act-pair · gpt-3.5-turbo

the word "Originally" and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

stakes

-0.79

atics

-0.74

 Roose

-0.74

oes

-0.73

owl

-0.71

 tears

-0.68

eer

-0.67

ifling

-0.66

rom

-0.66

lake

-0.65

POSITIVE LOGITS

Originally

1.14

 Originally

1.10

 Posted

0.86

ename

0.76

Filename

0.75

 conceived

0.75

 nicknamed

0.72

DEF

0.72

ãĤ´

0.70

 appre

0.68

Activations Density 0.005%

specific keywords related to the origins or initial history of something

the word "Originally" and its variations

No Comments

No Known Activations