INDEX

Explanations

years, specifically focusing on the year 2010

oai_token-act-pair · gpt-3.5-turbo

references to the year 2010

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

Magikarp

-0.80

ournals

-0.76

ques

-0.71

ess

-0.70

ournal

-0.70

phy

-0.68

amins

-0.67

esses

-0.67

vet

-0.66

heid

-0.65

POSITIVE LOGITS

å¹

1.08

 onwards

0.79

 census

0.76

acious

0.74

-'

0.72

ullah

0.71

assetsadobe

0.70

ford

0.70

é¾

0.70

 Tigers

0.69

Activations Density 0.022%

years, specifically focusing on the year 2010

references to the year 2010

No Comments

No Known Activations