INDEX

Explanations

phrases or words that suggest a physical force or impact

oai_token-act-pair · gpt-3.5-turbo

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

awaru

-0.68

 Palestin

-0.67

EMENT

-0.67

BILITY

-0.67

 Democr

-0.67

TAIN

-0.63

GMT

-0.61

BDS

-0.60

 VIDEOS

-0.60

 Bethlehem

-0.59

POSITIVE LOGITS

ashing

1.19

ogged

1.17

ojure

1.17

ipper

1.14

avier

1.12

amped

1.10

amps

1.07

umps

1.06

unky

1.06

ique

1.04

Activations Density 0.009%

phrases or words that suggest a physical force or impact

No Comments

No Known Activations